LED technology is a primary choice when users are concerned with energy consumption, controlled wavelength, and operating life. The latter is indeed a great advantage for LEDs in comparison to other technologies, with operating life sometimes exceeding 50K hours. Unfortunately, this tremendous advantage comes with a caveat; the light output of any LED decreases over time and may, after a certain time, reach a level outside of the required standards and its own original design specification.

Moreover, the emitted wavelength also shifts over time. This phenomenon also brings the device within the realm of unacceptable conditions in regards to appearance, efficacy, and color spectrum emission. Together, these two time-shifting quantities are important metrics when it comes to defining the performance of an LED device. The standardization of the measurement methods for these quantities is therefore a necessity for the entire lighting industry. The most common standards used today are LM-80-15 and TM-21.

Industry Lighting Standards: LM-80 and TM-21 

The Illuminating Engineering Society of North America Testing Procedures Committee (IES TPC) developed the IES LM-80 standard. This lighting standard describes the measurement methodology for two key LED performance metrics. The LED light output, also known as the luminous flux, and the chromaticity, or correlated color temperature (CCT). Correlated color temperature is essentially a scale of how cool or how warm the light emitted appears. This scale, in Kelvin, ranges from about 1000K (very warm) to 12 000K (very cold). Below is an example of different sources of commonly used lights and their measured correlated color temperatures.

The LM-80 method focuses on testing procedures that include conditions, operations, and data collection methodologies. Like with any standard, the goal is to provide a reliable comparison framework of LED performance with devices produced by various manufacturers and tested in various laboratories. The latest version of IES LM-80 (2015) defines a test duration of 6K hours in order to collect enough data points for acceptable data extrapolation. 

Once the LM80 testing has concluded, the measurements are then used to determine a depreciation curve for the luminous flux. The data is extrapolated using the method described in the Technical Memorandum TM21 to determine the “Lumen Maintenance Life Projection (Lp)”. The endpoint of the extrapolation is typically 70% of the initial light output, noted L70.

Since 2008, the IES TPC has developed a series of “paired” documents for testing and projecting luminous flux for component level (LED packages, modules, remote phosphors, etc.) and final product level (LED lamps, engines, and luminaires). These paired documents include IES LM-80 and IES TM-21; IES LM-84 and IES TM-28; and IES LM-86 and IES TM-29. To learn more about these, check out this presentation. 

There are factors that must be considered when designing integrated LED fixtures, packages, or modules alone. In the fixture’s application and operating environment, the design of the fixture is critical, from the drivers and optics to the overall thermal management. 

The latter is often considered to be the number one challenge when it comes to designing a device that will perform to an acceptable or exceptional LED light output level for a required lifetime. Indeed the rate at which the LED light output decreases over time depends highly on the temperature at which the LED chip is maintained. More specifically, it is the junction temperature between the die (the part that emits light) and the rest of the chip that is usually recorded as it is the warmest of all parts of the device. The higher the chip temperature, the faster the LED light output will decrease. The graph below, from the light research center, shows such a trend. On top of this,  the difference in operating lifetime has been extracted from the curves.

The example graph above shows that a decrease of 11°C on the LED temperature means that the operating lifetime is prolonged by 25K hours!

In this context, computational fluid dynamics (CFD) helps the designer to numerically test and predict the temperature of the LED and its fixture, to determine the best LED cooling technique. Such thermal performance testing occurs way before a prototype is built and the LM-80 (Lumen Depreciation of a LED Chip) test is carried out.

The temperature of the LED  is directly dependent on the heat dissipation of the whole LED device, therefore the thermal performance aspect is of prime importance during the design process. The second part of this article focuses on how design engineers can use CFD in the cloud to numerically test their LED device design and predict such thermal performance.

Evaluating the Thermal Performance of Lighting Solutions in the Cloud 

Testing the thermal performance of an LED device through CFD simulation is a fairly straightforward process. Whether the device is to employ LED cooling through active or passive cooling, the main objective stays the same; ensuring that the LEDs are kept below the maximum temperature recommended by the manufacturer and identify, if needed, areas of improvement in order to lower the temperature further. 

The CFD simulation produces both quantitative (the temperature reached by each of the components for instance) and qualitative (distribution of the temperatures, hotspots, and recirculation of surrounding air) results. These types of results enable the engineer to make better, informed decisions on what to improve in their design. 

While heat is transferred through multiple mechanisms; conduction, convection, and thermal radiation, there are many ways for the design engineer to improve their design. For the conduction of heat through the material, choosing a more conductive material or grade of material can significantly help the dissipation of the heat away from the LEDs generating heat. The thermal conductivity of the printed circuit board (PCB) on which the LEDs are typically mounted has a significant impact on the heat dissipation of the whole device. It is also one of the lower conducting parts due to the dielectric layers and orthotropic/direction-dependent material properties. This part of the webinar delivered on thermal performance analysis of lighting devices provides such an example.

Designers can also look at decreasing the thermal transfer resistance between the heat sink and the ambient air by altering the heat sink geometry and more specifically the fins or pins. Increasing the overall surface area in contact with the surrounding flow is a common way to decrease this thermal transfer resistance. Lastly, the resistance can be further reduced by improving the radiative properties of the heat sink, by using darker colors and surface treatment like anodizing or powdering. All of these angles of attack are possible ways for the design engineer to ensure their lighting device operates under acceptable conditions.

Conclusion

Design engineers can use CFD simulation to test multiple design variations simultaneously, within a few hours or even minutes, thanks to the power of the cloud. The results bring valuable insights about the heat dissipation that can be expected from the design, and specifically the temperature of the LEDs. Such insights constitute a reliable and accurate source for making informed decisions and choose the right design for the job and the budget.

The IES standards LM-80 and TM-21 set an industry-recognized method for measuring the light output depreciation over time and calculates the operating lifetime of an LED device. Such light output depreciation rate depends to a significant extent on the temperature at which the LEDs are maintained. 

More information about SimScale’s lighting design evaluation and improvement capabilities can be found in this webinar.

Additional References to Learn More About LED Cooling and Thermal Performance: 

• https://www.ledsmagazine.com/leds-ssl-design/packaged-leds/article/16695947/illuminating-engineering-society-updates-led-testing-standards-magazine
• https://www.ies.org/lda/maintenance-just-as-important-for-leds/
• https://www.penglight.com/lm80-lm79-l70-and-tm21-what-are-the-differences-for-lighting/

The post Thermal Performance of Lighting Solutions: LM-80 and TM-21 Lighting Standards appeared first on SimScale.

The extraction of air should be evenly distributed across the 5 extractors within the room in order to remove any potential contaminants effectively and also reduce potential high velocities, above 10 m/s, in the ducting. A minimum flow rate of 2.6 m3/s is required for the blower at a rotating speed of 2500 rpm.

The main objective of this project is then to use CFD through SimScale to evaluate the flow rate distribution across the extractors and the efficiency achieved by the blower for a base design.

How an Industrial Blower System Can Be Set Up for CFD Analysis

The CFD analysis is performed on a 3D CAD model that represents the volume occupied by the air. This 3D geometry can be uploaded as any CAD file format to the browser-based platform (SimScale) and can also be imported from the 3D modelling platform Onshape. Both the extracting space and the discharge space have a face assigned with an atmospheric pressure boundary. The flow rate achieved by the blower at 2500 rpm will be calculated by the solver.

Cloud-Native Simulation for Industrial Machinery Manufacturing

Our latest eBook explores how cloud-native simulation is transforming the challenges of industrial machinery manufacturing into opportunities. Download it for free by clicking the button below.

Download eBook

Base Industrial Blower Design Results

The simulation takes about one hour until it converges to a stable solution. The results can be processed from a web browser within the SimScale Workbench. 

The picture above shows the distribution of the flow rate through the extractors. With each extractor cut through by a velocity map, the large difference (0.6835 m3/s to 0.3835 m3/s) of utilization between the leftmost and rightmost extractors is highlighted. The air in the extractor closest to the blower reaches a velocity above 10 m/s, which indicates an unsuitable design to satisfy the operating conditions stated above. The picture also shows low-velocity regions in the main duct, as well as visible recirculation zones where the air flows back towards the extractors.

On the blower side, some low-velocity zones past the leading edges of the blades can be observed. This indicates flow separation as the change in incidence angle is too large as the air flows along the blades. It, therefore, hinders the flow performance of the device.

With this base design, the blower capacity is 2.51 m³/s, which is a non-satisfactory condition. Consequently, a new design is proposed so that the flow is better distributed amongst the extractors and the blower produces the required flow rate of 2.6 m³/s.

New Proposed Design for an Industrial Blower

The pictures below show the changes in the industrial blower extraction system, with smoother transitions and large elbows between the extractors and the blower inlet. This is to account for the recirculation and low-velocity phenomenon that could be seen in the original design results. To evenly spread the extraction, the individual ducts from the extractors are merging at one single plenum.

The blower blades’ angle of attack is increased from 20° to 40°. The blades of the proposed design have less curvature as a result of the flow separation phenomenon observed in the base design.

Comparative Results: Original Design vs. Proposed Design

The simulation results show a significant improvement in terms of flow distribution across the extractors with very little variation in flow rate (max 0.07 m³/s difference). Top velocities are also lowered, with a maximum air velocity of about 9 m/s.

The velocity in the ducts is more evenly distributed, as seen in the velocity cut plane results shown below.

The velocity cut plane through the midplane of the impeller highlights that the improved blower design is capable of generating higher velocities than the base design and therefore achieves a larger flow rate or 2.825 m³/s at a rotational speed of 2500 rpm. The large flow separation on the upper surfaces of the original blades is removed.

Whitepaper: On-Premises vs. SaaS Simulation: A Comparative Look

This whitepaper addresses the difference between on-premises software and SaaS
solutions for computer-aided engineering, explaining how SaaS came to be and its
key benefits.

Download White Paper

Conclusion: Improved Industrial Blower Design

With CFD, evaluating whether a design meets a specific performance requirement has become a straightforward and easily compatible design step for the engineer. With a 3D CAD model and a set of known conditions, the setup replicates a working scenario and the results include velocity distribution, pressure, temperature, and other quantities used for performance assessment.

In this project, an industrial blower system design was numerically simulated. The results produced indicated that the flow rate achieved by the blower was lower than the requirements. The extractor’s utilization was found to be uneven, resulting in velocity above the recommended value for vibration and excessive noise prevention. This led to the proposal of a new design, which was simulated under the same conditions.

The new design included a higher angle of attack and less curvature of the blower as well as a redefinition of the ducting between the extractors and the blower inlet. The results were compared with the base or original design and proved that the new blower system design achieved the desired flow rate of 2.6 m³/s, and the extractor utilization was acceptable, with velocity values lower than 10 m/s.

The post Predicting the Performance of an Industrial Blower With CFD appeared first on SimScale.

Over the course of their lifetime, very often over the 50K hour mark, most will continue to emit light until it diminishes in such a way it provides notice to its users; time to replace the bulb.

Gauging the Lifetime of an LED

Due to this unique characteristic, a metric called the lumen maintenance, or lumen depreciation, is used to determine the lifetime rating of an LED light source. It is defined as luminous flux that an LED, or rather the LED performance, maintains until falling below a specified percentage.

For industrial applications, the most commonly used lumen maintenance is L70, which refers to the number of hours for which the LED light output has dropped to 70% of its original lumen level. In-depth research conducted by ASSIST (Alliance for Solid State Illumination Systems and Technologies) highlighted this value after having observed that most users hardly notice a decrease in light intensity, even past 70% of their original luminance value. 

Every lighting device manufacturer following industry standards aims at providing such a quality metric to its customers as a reputable and accepted standard in the industry. Obtaining such a value is essentially a combination of physical testing and extrapolation-based data thereon.

Physically testing LED performance, and ultimately its LED cooling strategy, for the entire lifetime (50K+ hours) would mean the light would need to be turned on 24/7 for 5.7 years. As this testing timeline innately lacks feasibility, the IES (Illumination Engineering Society) created the LM-80-20 specification.

The LM-80-20 Specification

This standard is an approved method for testing LED lamps, arrays, and modules. It aims to determine the lumen depreciation characteristic and report the results over a period of 6K hours to 10K hours. A respective device is tested at 3 junction temperatures; a temperature that the manufacturer selects, 55°C, and 85°C.

In this test, the junction temperature refers to the temperature of the LED die inside the LED lamp as it is considered an excellent indicator of the quality of the thermal management of the whole system. Indeed, the higher the temperatures, the higher the luminance depreciation is, as shown in the graph below.

Once the LM-80 testing has been carried out with a minimum of 6K hours, the collected data points are used to extrapolate the lumen maintenance life to determine the L70 value. This method of extrapolation uses an exponential curve fit which is described by the TM-21 specification and is now recognized as the predominant standard for extrapolating luminous flux values. With this extrapolation method, the time to drop to a certain level of luminous flux can be calculated and would typically range from 36K to 60K hours.

As mentioned previously, the lumen maintenance over the lifetime of the LED will largely depend on the operating temperature of the LEDs, and therefore an efficient thermal design, aka LED cooling, is paramount. In this context, design engineers must ensure that their product design maximizes heat dissipation so that the LED temperature is as low as possible under the given operating conditions. While physical testing for thermal management is common, the cost and labor that come with it make the use of numerical simulation a viable addition for faster and more reliable predictions and optimization.

LED Cooling Testing With Thermal CFD Simulation

Numerical simulation, in the form of a computational fluid dynamics (CFD) simulation, comes with a number of advantages when compared to physical prototyping and testing. With 3D modeling being recognized as a standard design tool, it is easy to use these digital assets created to form a digital prototype that can then be simulated under operating conditions. These include multiple LED cooling strategies; forced or natural convection, liquid or air-cooled.

LED Cooling: Lighting Simulation Setup

The simulation setup can be easily carried out with minimum investment in time, cost, and training, over a vast array of designs that can be simulated in parallel. The most efficient design in terms of geometry, size, layout, and material selection can, therefore, be easily identified. In physical testing, it is typically very difficult to measure the LED die temperature by direct mechanical means, and it can often lead to inaccurate results. 

Some manufacturers use some approximations, with the solder point temperature (the temperature of the thermal pad on the bottom of the LED lamp), the thermal resistance between the LED junction and the solder point of the LED lamp, the current, and the voltage through and across the LED lamp. With the use of CFD simulation, this junction temperature can be easily assessed through a surface temperature probe. In addition to that, numerically simulated prototypes will give an accurate representation of the temperature distribution across the entire system, highlighting hotspots, excessive pressure drop, and flow recirculation zones. This kind of qualitative result gives explicit information that leads to better performance and efficiency. For instance, recirculation flow regions can be identified and design changes in certain areas can be made so that such effects are mitigated.

To find out more about the technical details of our new CHT solver, download and read the white paper below.

Download White Paper

Case Study: LED Cooling Through Liquid Cooling

This LED cooling project illustrates a typical use of thermal CFD analysis as a temperature and water cooling performance prediction tool for a 2x160W COB Series UV LED. These devices are primarily used for curing applications, as they emit light at specific wavelengths. The main objective here is to determine the temperature of the boards and make sure they stay below the recommended temperature of 40°C with the suggested water cooling block design running with 0.2m/s inlet velocity. 

The setup starts with importing the 3D geometry that represents the COB LED and its proposed design for the cooling block onto the SimScale workbench. The exploded view below shows the COB LED and its components.

The two COB LEDs are placed side-to-side on top of a steel cooling block. The cooling block is split in half horizontally through its middle plane so that a channel can be machined on both sides. An O-ring is placed around it to prevent any leakage. The blocks are held together with five cap screws. The COB LEDs are also screwed to the top cooling block with set screws.

The material properties for each component can then be assigned for each volume of the geometry followed by the flow conditions that represent a specific operating condition of 0.2m/s, and water at 20°C.

For LED lamps, it is known that about 80% of the power received is converted to thermal energy, therefore, 129W is assigned to each LED chip. This project uses the thermal resistance network definition to simulate the heat resistance in all directions from the LED chips (see more about thermal resistance network here).

With these known values of inlet mass flow rate and thermal power, an analytical solution for the temperature at the outlet can first be determined with the following equation: 

$$ \dot Q=\dot m Cp \Delta T $$

With a mass flow rate of 2.87e-2 kg/s, a water specific-heat of 4180 J/(kgK) at 20°C and the previously stated power source of 129W, the outlet temperature is expected to be 22.1°C.

LED Cooling Results & LED Performance

Benefiting from the power of cloud computing, the simulation runs take 60 minutes to complete. The temperature distribution can be visualized over the device and particularly at the faces between the LED board and the LED cooling block. One can easily identify the hottest spot on the LED boards, especially towards the center, where the two boards touch. The temperature then gradually decreases through the block. The surface of contact between the LED board and the cooling block shows a temperature of 36.15°C.

For the flow region, the computed temperature at the outlet is 22.45°C, whereas the temperature analytically calculated is 22.1°C. The velocity streamlines indicate the overall flow pattern of water inside the cooling block cavity.  The velocity increases at each elbow and at the reduction in the cross-section area between the connectors and the cooling block cavity. A recirculation zone is clearly identified at the output connector with streamlines circulating back towards the cavity of the block. 

This unwanted phenomenon, where energy is spent towards flow circulating in the wrong direction, means that some improvement could be made in either the cooling block cavity or the connector, smoothing out the exit cavity or minimizing steps in the geometry.

The pressure drop across the device is 740Pa and corresponds to 0.021W of hydraulic power at 0.2m/s inlet velocity.  This value would, therefore, help to select a suitable pumping device for the job.

LED Liquid Cooling Conclusions

This LED cooling CFD project gives a practical example of a CAD design that can be numerically tested as a digital prototyping method. This type of simulation can predict the junction temperature of a COB LED, liquid-cooled with a water-cooling block. The temperature values give important indications when it comes to assessing the overall lifespan, reliability, and lumen maintenance of the device.

The LED cooling results show that the temperature at the contact face between the copper LED board and the cooling block has satisfactory cooling conditions at 36.15°C (temperature max 40°C recommended).  The temperature could potentially be lowered, and the design overall improved, by using a more thermally conductive material for the cooling block, for instance. This would improve further the lifespan and lumen maintenance of the COB LEDs. Improvement in hydraulic power consumption can be made in order to reduce the energy consumption of the device by eliminating recirculation zones. 

To read more about electronics and LED cooling with SimScale, check out these recent blogs:

• Cloud-Based Battery Thermal Management for Electric Vehicles
•  Electronics Cooling Project: LED Spotlight
• Battery Cooling: Challenges & Solutions
• 5 Ready-To-Use Electronics Cooling Simulation Templates
• Thermal Design Optimization for Better Electronics Cooling

The post LED Cooling: Liquid Cooling Thermal Management appeared first on SimScale.

Battery Cooling: What Are the Challenges?

There are many different rechargeable batteries available on the market, varying not only in energy and power density but also in production and maintenance cost, runtime, safety, reliability, and overall life cycle durability. These types of batteries include Nickel Cadmium (NiCd), Nickel-Metal Hydride (NiMH), Lead Acid, Lithium-Ion (Li‑ion), and Lithium-Ion Polymer (Li‑ion polymer). More information on their respective power and energy density as well as their main domain of application can be found on circuitigest.com and batter university.com.

Once a battery technology has been selected for an application,  there are a few crucial steps in the design of the battery pack itself. These steps include the overall electric design in order to achieve the right voltage, power, and energy in balance with the cycle life, reliability, and safety. 

The first step includes both the mechanical and structural design of the pack. The final design must be able to withstand a variety of vibrations, pressure, shock, and crush loads under specific conditions. 

Additionally, the thermal design aims to maintain the cells within a given operating temperature range so that their longevity, performance, and safety is guaranteed. This last design consideration, applied to Lithium-Ion (Li‑ion) is the main focus of this article.

Battery Cooling: The Importance of Thermal Management

Batteries, just like humans, are happiest when kept at room temperature, both for working and resting cases. Li-ion cells and other battery technologies such as NiCd and NiMH react and perform differently with respect to the ambient temperature. 

For this reason, many manufacturers chose the operating temperature for the battery specifications to be  27°C. A battery will have a better performance at an elevated temperature, but at the cost of a shortened lifecycle if the exposure is maintained over a long period of time. The graph below shows an example of a reduction in the voltage value with lower temperatures of a standard 18650 Li-ion battery.

On the other end of the spectrum, at low temperature, the capacity rapidly decreases with the temperature. For example, a battery will typically deliver at –18°C only 50% of its capacity at 27°C. This is because lower temperatures affect the battery’s electrochemical reaction rate.

Many embedded Li-ion battery packs found in high-end portable devices are required to perform under harsh environmental conditions, ranging from -40°C to +80°C, like in military grade sensors. In such cases, unique design strategies and consideration must be deemed to produce sufficient cooling.

Different types of cooling strategies can be used, depending on the operating conditions and cost of manufacturing and maintenance. They range from liquid cooling, which is a very efficient type of cooling, to air cooling, which is not as efficient but a cheaper option because it is easier to implement. Air cooling can be performed through mechanical means, by a fan for example, or simply rely on the density-driven flow of air at varying temperatures, also known as natural convection. 

Maintaining temperature uniformity throughout the entirety of the battery pack is also crucial for the engineer to look at during the design stage.  This notably impacts the overall efficiency and prevents early deterioration, safety issues, and in extreme cases, thermal runaway or even explosions. In other words, if there is a high-temperature range across the whole battery pack, then there will be a strong electrical imbalance and therefore the overall performance will suffer. 

The Use of CFD as a Numerical Prediction Tool for Thermal Management of Battery Packs

When designing battery packs and their respective casings,  manufacturers must assess the thermal performance of their designs under working conditions by building actual prototypes and run extensive series of tests, often over a long period of time, always at a high cost.  During these tests, a myriad of sensors and other measuring devices are used to accurately record temperature, pressure, and velocity values. These recorded values are then compiled to verify whether the batteries are kept within the manufacturer’s recommended temperature range.

As part of a design process, whether it is at the evaluation phase, development phase or even the validation of an existing design, the use of numerical prototyping through the use of 3D CAD and computational fluid dynamics (CFD) has been shown to alleviate the cost, effort and time implied with physical prototyping. 

With CFD cloud-based thermal tools available such as SimScale, numerical simulations can be easily performed from a standard internet browser, using the power of the cloud for safe, fast, and multiplied computational power as well as unlimited data storage.

Another advantage of numerical prototyping is that manufacturers can test multiple designs iterations, to identify the optimal layout that reduces both material and manufacturing costs, without sacrificing the performance of their product.

Our Case: Active Cooling Designs for LI-On Batteries

In this SimScale project, inspired by this article on cooling effectiveness, a CFD thermal analysis is used to predict the temperature of 30 commercial Li-ion 18650 cells under multiple casing designs and inlet flow conditions. The goal is to find the minimum cooling power so that the cells are kept below 40°C.

This analysis uses the model created by user “Nilesh” on GrabCAD and represents a 10s3p ( 10 rows of 3 cells) of Li-Ion cell battery pack and a Battery Management System “BMS” represented by an electronics unit board at the extreme of the battery pack. The first proposed design of the casing hosting this battery pack consists of an 80mm cylindrical air intake, a straight 400mm rectangular mid-section, and vertical outlets vents. The complete model is shown below after being imported on the SimScale platform.

In this project, both solid conduction and flow convection heat transfer mechanisms are modelled using our conjugate heat transfer solver. The thermal properties (conductivity, density, and specific heat) of the battery cells themselves, the BMS board, and the battery will significantly impact the temperature distribution within the cells.  

The gravity direction was enabled. Material properties were then assigned to the components of the geometry. The cavity of the casing was defined as a volume of air. The solid components are the battery cells, to which the thermal properties were previously calculated and based on the properties of nickel. The battery spacers were assigned with ABS and the BMS board as PVC.  For this simulation, the walls of the casing were considered adiabatic which means that the heat exchange at these walls wasn’t taken into account.

Like in every numerical simulation where partial differential equations (PDE) are solved, a number of known conditions of the flow and power are required for the simulation to be run. The known operating conditions were assigned as velocity and temperature at the inlet and pressure at the outlet so that they are open to the air like in real-life conditions. Multiple air velocity values were simulated simultaneously to find the minimum value leading to a battery cell temperature below 40°C. The starting point chosen is 4m/s  at 23°C.

This project assumes that each cell generates 2.75W of thermal output power. This value was previously determined in this article which uses the same type of cells and a similar setup. The simulation is then ready to be run, and takes about 1 hour to complete.

Results and Design Decision 

• Why is this design not efficient in achieving a safe operating cell temperature?  
• How can the geometry be optimized to cool the cells better?
• Is there a potential to improve the design?

Evaluating the cooling effectiveness of the casing, one can naturally start looking at the overall temperature distribution. With an inlet velocity of 4m/s and shading the results by temperature, we can see that the peak value was 128C. This is 90C above the maximum recommended temperature of 40C and is currently a design failure.

These two result values indicate that either the design is poor for sufficiently cooling the cells or the inlet velocity is too low. 

As mentioned previously, simulations with different inlet velocities are run simultaneously in order to obtain a value for the minimum inlet air velocity required. Therefore, the maximum temperature value for the battery cells for each air velocity value can be plotted and we can easily visualize which operating condition is OK.

One can note that only very large inlet velocity values ( +35m/s ) will result in an acceptable temperature at the cells, and in practice could hardly be achieved by a standard fan/blower.  The second graph denotes that the airpower input grows exponentially with the velocity values. This means that to cool down the battery pack by a couple of degrees in the higher velocity values, a large amount of power is required.  

This can be explained by looking at the air velocity at the midplane, at 5m/s, and 10m/s velocity inlet.

It is well known that parts are more effectively cooled when the air passing them is faster. In this case, as the air would always choose the path of least resistance from the inlet to the outlet, one can see that higher velocities don’t encounter much of the cell’s hot surfaces and therefore very little cooling effect takes place.  

The central row of cells is with relatively low velocities (<1.5m/s). This gives an explanation as to why such high velocities are necessary to cool down the battery cells below the acceptable temperature. One can conclude that such gaps around the battery cells contribute to a very poor cooling capacity of the casing design.

What Changes Should Be Made to the Proposed Battery Cooling Design?

We learned that the central row is too hot, and we also saw that a lot of air simply passed around the outside of the cells, taking the path of least resistance. Based on these observations, two new designs are proposed. They both include a much narrower passage on the sides and upper part of the battery pack.  The difference between the two designs is the upper gap size. The new designs have a smaller inlet diameter of 50mm (previously 90mm).

These designs can be simulated simultaneously using the same parameters (material and boundary conditions) and their results processed to be compared.

Battery Cooling Design Iteration Results Comparison 

Comparing the maximum temperature of the battery packs in each design, at different inlet velocity, design 2(with the narrowest passage)  achieves the acceptable cooling performance, at an inlet velocity condition of 5m/s. The maximum recorded temperature is just below 40°C, at 39.09°C, the lowest temperature is 28.35°C (range is 10.74°W). 

The maximum temperature comparison chart shows a significant difference between the original design and the new design iteration, highlighting a better cooling performance for the new casings.

When looking at the required air input power for each design,  we have to be careful when reducing the air space too much as it means that the power necessary to move the air through the system will be prohibitively expensive.

Looking at the satisfactory design and operating conditions of 5m/s, the corresponding air power required is about 9.81W. With this velocity and wattage, it is easy to select the correct fan, so that we would choose the fan that can operate most efficiently.

In comparison, design 1 would reach a similar temperature (~39°C) at the battery cells for 15m/s (36.79W). 

Conclusion

Maintaining battery packs within a specific temperature range is an essential aspect to guarantee the lifespan, reliability, and safety of such components.

The cooling performance assessed in this project includes: 

• The temperatures reached by the battery pack cells. 
• The power necessary in order to achieve such a temperature. 

The importance of forcing the air to flow at high velocity between the cells has been highlighted in the proposed designs. These improvements include narrowing the gaps on either side of the battery pack, as well as the upper gap. This way, more air is forced around the cells and therefore heat transfer can occur at the cell surfaces.

The post Battery Cooling: Challenges & Solutions appeared first on SimScale.

Spending so much of our lives in buildings has made it necessary that the thermal environment and indoor air meet certain occupants’ standards. And actual official standards, i.e., ASHRAE 55, EN 15251, EN 16798-1:2019, ISO 7730, ASHRAE 62.1, have been put in place for architecture and engineering projects.

Comfort and air quality are the main topics for building designers, as they have a direct impact on wellbeing, productivity, and even more importantly, health and safety. From residential buildings, commercial spaces, schools, and offices where thermal comfort is a top priority, to factories, clean rooms, and underground parking garages where indoor air quality, contaminant removal, fire safety, and smoke management are obligatory, building designers know that heating, ventilation, and air conditioning (HVAC) is their best friend.

A recent news story about how important an HVAC system is that the Internet community found it hard to believe at first is that of a Malaysian cinema and luxury leather goods found covered in mold after a two-month lockdown due to the COVID-19 pandemic. The significant financial loss that resulted would have been avoided if the air conditioning hadn’t been turned off.

As unfortunate as this incident was, however, the negative effects on comfort and health that poor ventilation has are much more common and concerning.

What Influences the Thermal Environment?

The thermal environment is affected by temperature (of air as well as surrounding surfaces), humidity, and airflow rate. A satisfactory indoor air is characterized by comfortable room temperature, proper humidity, cleanliness, and freshness. 

As defined by ASHRAE 55 and ISO 7730, thermal comfort is “a condition of mind which expresses satisfaction with the thermal environment and is assessed by subjective evaluation”. In HVAC design, thermal comfort is measurable via predicted mean vote (PMV) and predicted percentage of dissatisfied (PPD) calculated for a given space. More about them you can read in this blog post: What Is PMV? What Is PPD? The Basics of Thermal Comfort.

The following simulation project aims to serve as both an example case and a guide for investigating the thermal environment and indoor air quality for a building.

HVAC Case Inspired by Cité du Design in Saint-Étienne

This project is an example of how useful computational fluid dynamics (CFD) simulation can be for early-stage HVAC design. The CAD model used is an exhibition hall, inspired by the Cité du Design building located in Saint-Étienne, France. 

The building comes with several challenges that HVAC engineers had to overcome in order to ensure proper indoor air. Its walls and the main part’s ceiling are entirely made of glass and steel tiles. In the absence of a suspended ceiling, the classic approach of placing diffusers couldn’t be undertaken. Hence, the engineers had to come up with a bespoke solution in order to guarantee an acceptable indoor air quality and a suitable level of thermal comfort for the occupants. 

This also means that the amount of solar radiation that the exhibition hall is exposed to is significantly higher than a classic space surrounded by opaque walls and ceiling. This radiation effect will play a strong part in the amount of heat that is introduced to the HVAC system, and, in a room filled with many objects and complex shapes, it is generally a very difficult task to quantify with traditional methods.

Traditionally, HVAC design processes include selecting an air change rate (ACR) corresponding to the type of room, defining a number of diffusers, and calculating the spacing between them so that their throw and drop values would suffice to cover the specified areas. The air intakes would also have to be conveniently located to prevent stagnation of air, excessive recirculation, and hot spots. 

While defining an ACR is still relevant in the present case, the classic approach of selecting and locating diffusers from a generic manufacturer couldn’t be applied here and therefore a unique design approach had to be undertaken.

The exhibition hall is thus equipped with floor diffusers, and air intakes have the form of 3.8 meters-high chimneys, located on the four corners of the rectangular room. In the summer conditions scenario presented below, the cold air stream would be inserted from the diffusers, and then the intakes chimney would collect the heated air. 

The CAD model simulated also includes a simplified representation of the ducting system placed between the air intakes and the floor diffusers. A cooling unit and a blower effect can be simulated by simply assigning certain zones of this ducting system to a momentum source (a fixed velocity) and a power source. 

This simulation strategy brings several advantages when assessing the thermal environment and indoor air quality. Firstly, the cooling requirement for the whole system could easily be defined, including the power required to cool down the outdoor air introduced in the exhibition hall. This cooling power requirement, coupled with the mechanical ventilation power, is important when qualifying a building for green certifications such as LEED or BREEAM. Secondly, simulating the entire duct network would give interesting clues on the utilization of the different diffusers and air intakes. In other words, how even is the flow distributed among the diffusers and the air intakes, and does this impact the temperature distribution and therefore the thermal comfort of the exhibition hall’s visitors? Lastly, the closed-loop airflow will highlight with a pressure drop that is essential to quantify when assessing the power requirement for such a cooling system.

CFD Simulation Steps

The simulation starts with importing the 3D model of the exhibition hall and its ducting system. The model also includes the major furniture pieces that an exhibition hall can host—podiums, signposts, desks, chairs, separation walls, and freestanding posters.

The hall is 15x30x4m, and the ACR is selected to be 5, which defines the amount of air to be blown into the room to about 9500m3/h, with 500m3/h of outdoor air introduced. As with many HVAC designs, the excess air will escape through various openings, typically through the doors or smaller gaps in the ceiling joints. 

The diffusers located on the floor are imported as simple block volumes and the simulation setup will enable us to assign these volumes as perforated plates, with a specific free area ratio. 

This is a very useful feature brought by the online CFD platform used in this project (SimScale), allowing it to work with simple geometries without compromising on accuracy and relevance. The number of diffusers is 16 (four rows of four diffusers ) so that, for the given overall flow rate,  the velocity at each individual diffuser is not higher than 0.8m/s, as recommended by the ASHRAE 55 standard.

The walls, ceiling, and floor of the exhibition halls are defined for the simulation so that they take into account a certain value for thermal insulation, called the transmittance value (or U-value), the outdoor air temperature (30°C/ 86°F) and the external convective heat transfer coefficient (how much heat can be transferred at this surface). Last but not least, a value for diffuse solar radiation is assigned for the ceiling only, as the simulation replicates a worst-case scenario of a summer day at noon, with a typical value of 713W/m2.

The presence of large transparent surface areas in the model will make the radiation component a very strong heat transfer mechanism, a factor very challenging to approximate without CFD simulation. 

To evaluate the cooling requirements for this hall, a starting value of -6800W (negative because heat is removed from the system) is set. This is an analytically calculated value, without the radiation component. This cooling value can be then readjusted for a new simulation run after consulting the temperature and thermal comfort results. The thermal comfort metric used for this project is the predicted mean vote (PMV), which gives scales from -3 (occupants feeling cold) to 3 (occupants feeling warm). It is a sensation value based on several factors including the temperature, airspeed, humidity, mean radiant temperature, metabolic rate, and clothing of occupants.

Lastly, some quantities such as temperature and velocity at particular locations (center of the room, inlet ducts, and air intake chimneys) can be recorded throughout the simulation iterations so that the convergence of the computation can be confirmed, ensuring that the results are accurate and the numerical stability of the solution is reached.

Results

The right amount of cooling power necessary to maintain an acceptable thermal comfort value was found to be -40000W, and the PMV contour maps are shown below. 

Due to the specific arrangement of furniture and diffusers, one can observe lower PMV values above the diffuser locations because of the higher velocities at the diffusers impact the PMV values negatively. Overall, the PMV values are within the recommended range specified by ASHRAE 55.

The temperature contour map at 1.5m height indicates hotspots in the center of the hall with regions about 1°C above the average temperature of 21.65°C at this height. Cold spots corresponding to the location of the most utilized diffusers are mainly located in the corners of the exhibition hall, close to the air intake chimneys.

In practice, the opening of diffusers can be individually changed in order to achieve higher temperature homogeneity.

The mechanical power required to maintain the previously defined air change rate can be quantified through the pressure drop across the blower zone shown below and the velocity assigned to this same zone.

The pressure drop observed is about 130Pa and the flowrate is 2.64m3/s which gives a mechanical power for the blower of about 243Watts.

Conclusion

Testing the thermal environment and assessing the performance of an HVAC system through CFD simulation brings many valuable insights into the design process, especially with bespoke designs, where the configuration of a space is unique and therefore traditional HVAC design methods cannot be employed. 

One can easily quantify the cooling and mechanical energy expenditure of the building under specific conditions, an essential metric to select the right HVAC equipment and qualify any new construction for a green building label. 

The project described in this article demonstrates that such CFD tools contribute not only to determining this value, but also confirming that an acceptable thermal comfort of occupants is met. The complexity brought by furniture and other obstacles in the room resulted in uneven temperature distribution, even if the PMV values for thermal comfort were within range according to ASHRAE 55.

The post Thermal Environment and Indoor Air: Guide for Building Designers appeared first on SimScale.

Recently, engineers have to comply with mandated standards for the accelerated wind velocity pedestrians face from constructed buildings and structures when meandering down busy high streets (e.g., Lawson, NEN 8100, etc.). Additionally, environmentally-friendly measures to improve factors like air quality are also being increasingly employed. With more and more requirements engineers have to adhere to for their designs, engineers need to think critically on how to tackle multiple issues with one solution. In the case of air pollution and pedestrian wind comfort,  what is the best way to kill two birds with one stone? We’re glad you asked; by planting trees! 

How Do Trees Affect Pedestrian Wind Comfort?

In addition to converting carbon dioxide into oxygen and working to eliminate pollution and airborne toxins, trees also hinder the effect of wind as it passes through their leaves. Of course, trees come in a variety of shapes and sizes and are characterized by vastly different leaves, and wind velocity is merely decreased but not fully stopped as it passes through their branches. Naturally, it tends to come down to the density of a tree’s leaves, and the higher the leaf density, the greater resistance against the airflow.

With this in mind, we can assume varying types of trees offer more or less resistance to the wind flow. For example, the silver birch tree has less air resistance than the plane tree, based on it’s calculated leaf density. This ability to allow fluid (air) to pass through, known as porosity, is measured through experiments (see sources [1] and [2] ) and the results are then defined through a leaf area index.

When trees with a large leaf area index, offering the greatest resistance to wind, are placed in areas where fast airflow and strong wind effects (see upcoming webinar for more explanation on this topic) are present, they offer a great mitigating solution and therefore contribute to better wind comfort and safety for all the passersby.

How to Simulate Different Types of Trees: What Is Porous Media?

With SimScale’s new porous medium feature, users now have the possibility to model different families of trees. With built-in tree models based on a library of tree species, it is easy to set up a tree with the correct wind hindrance properties, and assign volumes or faces on the model that correspond to where the trees are located. Alternatively, users can also input specific porosity and pressure drop value to populate the parameters of any tree.

What Are the Benefits of This Feature?

Implementing the effect that trees have on the wind flowing through an urban environment is a great asset for our lattice Boltzmann method (LBM) toolbox. Architects, urban landscape managers, local councils as well as anyone in the general construction sector can now observe and assess the effects of trees in their urban landscape models.

Our Case: A Bustling City Center 

In our project, we used a fictitious model of a city center that features trees in parks, along streets, and around avenues to showcase our latest LBM feature: porous media. This feature can assist engineers in simulating air flowing through trees in a multitude of different scenarios.

In this project, we will compare a model with trees and a model without trees to illustrate the impact of this feature.

Simulation Setup 

The CAD model imported for this wind simulation represents a typical American city center with streets, pavements, terraces, and parks. It includes the 3D models of trees, in the form of different volumes, placed at a specific location along the streets and in the parks.

By selecting our LBM type of analysis, we essentially placed a virtual wind tunnel over this imported 3D model. This wind tunnel box, also known as the external flow domain, was the base volume in which the airflow was simulated. The size of this box was automatically calculated to fully capture the airflow on the city model.

This box can then be oriented to simulate different wind directions. For our chosen scenario, we were interested in observing what would happen if the wind direction was aligned with one of the main streets and facing the park at the center of the model. We defined the wind profile that entered the virtual wind tunnel with a typical 10m/s at 10m height basepoint.

We then set the realtime of the simulation to 300s, which was enough for the flow to go through the wind tunnel at least 3 times until the transient solution was fully established. We can use the advanced feature modeling items within the simulation tree to define a porous object region. We can either use the Darcy coefficient definition or the tree definition. 

In the popup window, the type of tree can be selected from a predefined list or by creating a custom tree type. As mentioned above, the porosity of trees is defined by their respective leaf area index, which can be inputted autonomously. In this scenario, we have selected a plane tree to be assumed for every tree within the model.

Lastly, we defined which areas the results would be calculated. Generally, these areas include a plane, located at 1.5m above the ground, the standard height for assessing pedestrian comfort. This is fully adjustable. 

Simulation Results 

The comparison between our simulations, one with the tree porosity feature and one without, shows significant differences in the flow pattern and wind speed at the pedestrian level (as shown in the pictures below). On the model not including trees, one can observe that the wind flowing from right to left along the main street (horizontal), maintains a speed of about 4 m/s until it reaches the middle of the intersection.

This typically means that according to the land Beaufort scale [3] and comfort criteria [4] such as Lawson, pedestrians could potentially feel uncomfortable sitting for a long period of time on terraces and benches at this location. Additionally, typical wind effects can also be seen, and in this case most notably the cornering effect where the air accelerates at the corners of buildings. 

When visualizing the results of the model including trees, one can see major differences, both in terms of wind speed and wind pattern. First, the wind intensity rapidly decreases as the flow enters the street on the right-hand side, restricted by the first rows of 4-5 trees on both pavements of the street.

With trees along the pavements and in parks, the mitigation of the cornering wind effect is very significant, cutting down the acceleration of the air around corners.

Porous Media Conclusion

Simulating the effect that porous zones have within urban environments is an increasingly essential tool for accurately assessing the actual wind speed and pattern at a pedestrian level. In this case, the cornering effect was clearly attenuated by the trees located on the pavements and in parks. In a previous article that explains how we simulated the Insight Partners’ building in NYC, another wind effect called the downwash effect was shown to be successfully mitigated by the presence of trees. Using SimScale, users can assign porous zones to tree models through a straightforward setup and benefit from even more accurate results for wind comfort simulations.

Additional Resources on How Our Porous Media Feature Can Help Your Wind Comfort Simulations:

• March 2020 Product Update From SimScale
• Advanced Modelling

The post How to Model Different Types of Trees with Porous Media appeared first on SimScale.

In this article, we will examine how a digital replication or CAD model of the Vessel in New York City can be evaluated for pedestrian wind comfort using CFD, and more specifically wind comfort analysis from SimScale.

Wind Comfort Analysis from SimScale

In 2019, we unveiled our new pedestrian wind comfort module that allows users to perform multi-directional wind analysis. Through creating a visual wind comfort map showing the different degrees of wind comfort—from blue (comfortable for long sitting) to red (uncomfortable), users can now evaluate and confer with different wind comfort standards around the globe including Lawson, Davenport, and NEN 8100. The animation below illustrates the results of a wind comfort simulation, where areas of discomfort appear in red around corners of buildings, as well as in the street where the skyscrapers have narrow streets between them.

This map provides a powerful plot to base design decisions on, enabling architects, urban development teams and municipalities to prevent issues that might arise after the project’s completion.

The Vessel, Hudson Yards, New York City

The Vessel in Manhattan, New York City is a newly built attraction for tourists and architecture enthusiasts alike. As part of the Hudson Yards Redevelopment Project, it stands 16 stories tall and is comprised of 2.5K steps in a captivating honeycomb-like shape. The total funding needed for the construction of the Vessel came to around $200 million.

This SimScale project simulates the airflow and ultimately determines pedestrian wind comfort around the famous Vessel structure in New York City. The coloring of the flow lines represents the respective wind velocities, where blue is slow and red is high speeds. This can be seen in the two images below. The results, produced by a Lattice Boltzmann method (LBM) simulation on the cloud-based platform SimScale, are a valuable asset when designing such structure and integrating it into an existing environment.

One can note that the airflow accelerates as it enters the structure through the openings, a large part is diverted upward, as a result of a building located right behind it. In addition, the presence of this building in the immediate route of the airflow generates recirculation zones, where the air rotates down its facade to flow back up at the rear of the Vessel structure. This highlights a zone of potential issues for the comfort and safety of pedestrians on the ground but also at different heights such as the terrace shown as a recess in the facade. An aerial view of the Vessel and its surroundings can be seen pictured below:

This project exists as a great demonstration of SimScale’s wind comfort analysis abilities, as well as the level of CAD model sophistication that is able to be meshed and simulated on a cloud-based platform. For more information on this project and the Vessel itself check out this video from The B1M:

More Wind Comfort Resources from SimScale

• Lawson Wind Comfort Criteria: A Closer Look
• Wind Comfort Criteria: Lawson, Davenport, and NEN 8100
• How to Analyze Pedestrian Wind Comfort with SimScale
• Sustainable Wind Engineering: The Stockholm Royal Seaport Project

The post How Can I Use CFD to Evaluate the Hudson Yards Vessel? appeared first on SimScale.

While there is no all-encompassing wind comfort criteria standard, there are many standards existing today to aid assessment of predicted wind environment for the built environment, by providing parameters of what should be achieved to stay within favorable conditions. To note the standards used widely by engineers around the globe, Lawson, Davenport, and NEN 8100 are among the internationally recognized.

In this article, we will focus on Lawson wind comfort criteria, and extrapolate on how even within this comfort standard there are further denominations to accommodate different variations of threshold wind speeds. Along with this, we will provide visual simulation examples from the SimScale platform to illustrate the different findings.

This paper addresses the topic of pedestrian wind comfort, from origin and
definition to wind comfort analysis, criteria, and example case studies; all meant to
form an in-depth understanding of the field.

Download Whitepaper

What Are the Lawson Wind Comfort Criteria?

The Lawson criteria are defined by the probability of one particular location to see wind speed higher than a certain speed. These speeds are measured at a particular height which is usually between 1.5 m and 1.75 m depending on the local authority rules. In simpler terms, the Lawson criteria set threshold wind speeds, and then dictate the probability of wind speeds exceeding that threshold.

The different wind speed threshold values, as well as the probability values, make the level of comfort for pedestrians. They usually correspond to an activity that would be able to be achieved in an acceptable manner, such as sitting, standing, walking fast, etc. The probability is calculated using statistical weather data. This statistical data is obtained from a year-round data collection of wind speed and frequency in 4 to 36 directions.

Starting with the highest threshold speed (the most uncomfortable or unsafe condition), the probability at each specific point is calculated, and if the probability is less than the one stated by the category, then the category velocity range is satisfied (and the wind velocity deemed safe for pedestrians!). The calculation continues with the following threshold speeds until the fulfillment isn’t met anymore, which means the probability is higher than the one set by the category. This means that this specific point has its wind comfort criterion set to the last fulfilled, where the probability was lower than the one set. This computation is made for each point, and for each direction. At each point, and for each direction, the threshold speed is scaled by an amplification factor computed from a combination of CFD results and meteorological data. This meteorological data takes into account the terrain type in each direction; they are associated with a factor value that represents how the wind is slowed down by obstacles like buildings or trees.

Within the Lawson standard, there are subtypes to accommodate different cultural perceptions, geographical locations, and local authorities. The wind speed thresholds used, as well as the probability values for each threshold, make up the type of Lawson criteria used. The different subtypes of the Lawson criteria (LDDC, 2001, 1970 etc.) will have different probability values and wind speed thresholds.

General Lawson Criteria

In its original form, Lawson wind comfort criteria are made up of five different categories that all use a probability of 2% as a fulfillment value.

The “Uncomfortable” (red/E) category indicates the most undesired areas, where the value of the speed is higher than 7.6 m/s, is likely to be more than 2% of the time. This category is usually quite rare, and as seen below we usually only see a small number of such zones on a wind comfort map. If such zones are used by pedestrians or cyclists, action should be taken in order to alleviate these unwanted conditions.

The category “Walking Fast”, shown in yellow, corresponds to points where the likelihood of seeing a wind speed higher than 7.6 m/s is likely to be less than 2% of the time. Basically this is true for all lower categories as well. So here the wind requirements of this category are met, but not those of categories A, B, and C. These points would indicate areas of concern depending on the intended use of the area (i.e., an outdoor dining restaurant would not work in these conditions). Typically, there is a threshold (linked again to the local authorities and wind standard) that is a “good quality” if the criteria of the designated usage are met, “acceptable” if it’s one category worse and “unacceptable” if it’s worse two or more categories.

The following categories, shown in green, light blue, and blue, represent points where the speeds higher than 5.3 m/s, 3.6 m/s, 1.8 m/s respectively are less than 2% likely to be observed. These points indicate normal conditions that pedestrians would most likely find comfortable. Under the given activities.

Lawson LDDC Criteria

The Lawson LDDC wind comfort criteria is a subset made of 6 different categories; the safest 5 categories use a probability of 5% as a fulfillment value. The “Unsafe” category, shown in red, represents areas with the likelihood of experiencing a wind speed higher than 15 m/s is more than 0.022%. Should some passage zones be classified with this category, safety measures must be put in place to reduce risks of accidents.

The following category, corresponding to “Uncomfortable” zones, shown in orange, corresponds to points where the likelihood of seeing a speed more than 8 m/s is higher than 5% of the time. Depending on the usage of such zones by pedestrians or cyclists, adequate measures should be taken.

The next categories; “Walking” ( D ), “Standing” ( C ), “Occasional sitting” ( B), and “Frequent sitting” ( A ) represent points where speeds of more than 8 m/s, 6 m/s, 4 m/s, 2.5 m/s respectively are less than 5% likely to be observed.

Lawson 2001 Criteria

The Lawson 2001 wind comfort criteria use categories ranging from A to E with the addition of two extra categories, both called “S”.

The most undesired “S” category, called “Unsafe all” and shown in red, is represented by areas likely to experience more than the 20 m/s threshold speed more than 0.023% of the time. Second to this category, called “Unsafe frail”, is showing areas with the potential to experience more than 15 m/s wind speed more than 0.023% of the time. This is the orange category.

The “uncomfortable” category, in yellow, shows the locations where the likelihood of seeing a wind speed higher than 10 m/s is higher than 5%.

The categories from A “Sitting” to E “Uncomfortable” represent zones where speeds higher than 10 m/s, 8 m/s, 6 m/s, 4 m/s, and 2 m/s respectfully are less than 5% likely to be observed.

Conclusion

Widely used in the AEC industry, the Lawson wind comfort criteria is recognized as an effective method when it comes to assessing pedestrian wind comfort for urban environments and cityscapes. It uses different threshold speeds and calculates probability given by the CFD results, the meteorological and terrain information, and the wind rose data. These make different levels or categories of comfort for pedestrians, they correspond to either a state (fail, unsafe), feeling (uncomfortable) or an activity (outdoor dining).

Based on the location, wind environment, and landscape of where engineers would like to build or expand on a design, these subsets of the Lawson criteria can help engineers make better and more specific design decisions. Here at SimScale, we offer different wind comfort analysis tools and standards to better fit your evaluation needs, and investigate pedestrian wind comfort in any location around the globe.

More Wind Comfort Resources from SimScale

• Wind Comfort Criteria: Lawson, Davenport, and NEN 8100
• How to Analyze Pedestrian Wind Comfort with SimScale
• Sustainable Wind Engineering: The Stockholm Royal Seaport Project

The post Lawson Wind Comfort Criteria: A Closer Look appeared first on SimScale.

The main challenge is to keep every electronic component within the operational limits set by the manufacturers, in order to guarantee a reliable and safe operation of the system. These operational limits can be of different forms, mechanical stress, fatigue, humidity, vibration, and thermal. The latter is the one which will be discussed in this article.

The Main Problem with this Application: Overheating

Thermal management is one of the most critical tasks when designing electronic systems and it is paramount that the designer integrates the prediction of critical temperatures as a part of the design lifecycle of their product. Accurately predicting temperatures distribution used to be considered an unattainable goal when dealing with complex and detailed models. These values could therefore only be partially evaluated thanks to built prototypes and full-scale experiments, worse case scenario estimation, and approximated design rules.

In this context, the development of computational fluid dynamics (CFD) and conjugate heat transfer (CHT) simulation for thermal management came as an alternative tool that would bring faster, more reliable, and accurate performance prediction. In addition to producing heat maps and temperature distribution plots, engineering simulation would give useful insights into whether a component’s junction temperature under operating conditions ensures high optical LED performance and sufficient lifetime. The results offered by numerical simulation allow designers to predict part and junction temperatures before any manufacturing has taken place.

How to Solve this Problem: Electronics Cooling Systems

In order to limit the temperature rise of electronic components, engineers need to keep them within the correct operating conditions and prevent overheating scenarios. Certain cooling strategies can be deployed by the designer, and the most commonly used are:

• Air-cooled by natural (also known as passive) or forced convection.
• Liquid-cooled through flow circulation of fluid with a pump through a piping system.

Being able to predict performance, evaluate certain design scenarios, and assess all working conditions before prototyping is of crucial benefit for any manufacturing company. This is particularly relevant for mass production and large series, where the material reduction and manufacturing costs are paramount. In fact, the cost of changing the design further in the production process increases the cost greatly in an upward trajectory, as shown below.

An undetected overheated component of a product that goes to market could result in a product recall to minimize potential dangers. We’ve seen this recently at the release of Samsung’s Galaxy Note 7 phone, where only after the device was brought to market it was discovered the battery could become overheated causing the phones to explode in some cases. The phones have since been recalled.

In the following case, we will demonstrate how engineers can evaluate products in the design phase from a 3D CAD model to accurately validate performance. Using CFD, users can experience great advantages over traditional methods including saving time, money, and obtaining deeper insights.

Our Case: Conjugate Heat Transfer (CHT) of an LED Spotlight

LED spotlight cooling is a suitable example where CFD can be used for optimizing a design in order to predict temperature distribution and therefore avoid components overheating. This overheating could result in catastrophic damage and failure to the whole system, adversely impacting not only the normal operation of the system but to a larger extent, the safety of human operators.

The model that is to be analyzed is a standard LED spotlight generating 9W of thermal dissipation power, these spotlights are fairly common in households and provide an energy-efficient solution as a lighting apparatus. The overall goal of the simulation is to verify that the heating elements of the LED spotlight design have their temperature kept under the 90°C/363.15 K operational limit. The results will provide an evaluation of the dissipation performance of the components of the LED light under natural/passive cooling conditions.

Simulation Setup

The setup starts with the import of the LED light’s CAD model, representing the most significant details of the geometry that will affect the flow distribution and the temperature profile of the device. The CAD model consists of a casing, a glass protector, a metal core, a supporting plate where the chips are placed, as well as electrical connector parts. For simplicity, only half of the whole model is used for this project and a symmetrical condition is assumed throughout. A rectangular volume enclosing this LED light model represents the air domain, this enclosure is created on the platform, once the geometry is imported.

This project is a conjugate heat transfer analysis, which solves both the conduction and convection heat transfer phenomenon. This way, the distribution of the temperature through the heat sink and the other solid components will be computed taking into account the convection from the airflow. Since this is a natural convection situation, the airflow is driven by the buoyancy effect generated by the heated surfaces.

Once this analysis is selected, the first step is to generate the mesh. Meshing is an operation that aims at generating simple basic elemental volumes of the different parts of the CAD model. These parts—which represent both solid and fluid volumes—are broken down into small elements where fluid governing equations will be computed by the solver. The mesh is defined with elements small enough to ensure that the geometry has been captured to a relevant level of detail and that the mesh density is high enough to accurately capture the flow and solid distribution of temperature, velocity, and other quantities.

Material Assignment

Benefiting from the parallel cloud computing, the rest of the simulation can be set up while the mesh is being generated. After the gravity is defined with the LED spotlight facing down, materials for the fluid and solid volumes can be defined. The casing is made out of PA material and the core made out of aluminum.

Boundary Conditions

With this simulation representing a LED spotlight in an open environment, the faces of the air volume are set as open boundary conditions, meaning that the air is free to go in or out of the domain. Since the simulation is simplified to half of the actual LED light model, the symmetry faces of both the fluid and solid parts are set to symmetrical boundary conditions.

The heating sources are assigned to the volumes that represent the LED chips made out of silicon, where three are assigned a power of 2W, and the one in the middle is set to 1W since it is cut in half by the symmetry plane.

Results

The solver is set up for 1000 iterations, and the top face of the 2W and 1W chips are set for monitoring the temperature as goes through these iterations, confirming when numerical convergence is reached. The temperature distribution on the solid part shows how well the heat is transferred from the heating element to the rest of the geometry.

A few observations can be made here. First, the heating chips have their temperature kept at 345 K, below the operational limit of 363.15 K; on average, 30 degrees higher than the PVC plate that they are attached too. This demonstrates the importance of the thermal interface between the chip and the board, which can sometimes be enhanced by applying some thermal paste. Second, the temperature over the rest of the components is quite homogeneous and kept within a reasonable range, i.e., lower than 300 degrees Kelvin/26 degrees Celsius, which is a temperature that allows bare hands handling.

The velocity profile as the air flows along the length of the LED lights shows the velocity ramping up as it catches heat from the higher temperature surfaces. Convective induction is formed at the extremity of the vertical surface, this effect is characterized by air streams joining behind the LED spotlight. This generates some recirculation zones at the upper face of the PA casing. The casing being hollow at its rear, some recirculation and stagnation zones can be denoted in this region, which results in heat being higher in this zone.

Conclusion

The goal of this electronics cooling project was to predict the temperature at the heating chips and heat distribution performance on the components of an LED spotlight. The CHT simulation was used to evaluate the flow and heat pattern of the specific design with specific material properties, geometry, and size, ultimately assessing the LED performance. The initial setup process was described for a specific load case and environment. The results highlighted the importance of the thermal interfaces between the different components in terms of heat dissipation, as well as the conductivity of materials in place. The fluid flow was then described to highlight some characteristic phenomenon of buoyancy-driven air flows present in natural convection electronic cooling case.

The results showed the surface temperature of 345 K at the heating chips which proves that the design was satisfactory in order to maintain the component within their operational limits of 363.15 K. This design could be further improved by, for example, improving the board conductivity by using thermal vias or increase its copper content.

Additional Electronics Cooling Resources from SimScale:

• White Paper: The Ultimate Electronics Cooling Guide
• 5 Ready-To-Use Electronics Cooling Simulation Templates
• How to Use CFD Analysis for Electronics Cooling
• Thermal Design Optimization for Better Electronics Cooling
• Testing the Cooling Performance of a Heat Sink for Different LED Packages

The post Electronics Cooling Project: LED Spotlight appeared first on SimScale.

This article illustrates how useful and accessible SimScale’s wind comfort analysis capabilities can be in order to tackle such challenges. This type of simulation has been developed to truly take advantage of the cloud-based computing power and offer accurate, transient and large scale results in a turnover time that is unprecedented in the industry.

Our Case: Stockholm Royal Seaport

In order to put our powerful platform to the test, we are going to look at the example of the Stockholm Royal Seaport urban development project. This recent expansion project that started in the early 2000s is set to create 12,000 new homes and 35,000 workplaces by 2030.

This smart grid development area in the capital of Sweden is a perfect candidate for a wind engineering simulation analysis, ensuring a sustainable, comfortable, and safe environment for the residents of the planned area.

This project is located over a large and exposed area of a semi-coastal environment, which makes wind analysis even more relevant due to strong winds coming off of the sea. Numerical simulation would greatly help architects, urban development managers, and structural engineers to not only spare time invested in designing layouts, elevations, and buildings but also in saving consequent financial investment in multiple resource-intensive wind tunnel tests.

For our case, we will be using the online CFD solution in the form of Lattice Boltzmann method (LBM) analysis provided by SimScale through its integration with Pacefish®. This method (LBM) has been largely developed for this type of wind simulation and, as opposed to traditional steady-state CFD analysis, to easily capture any transient effect of wind passing between buildings, under bridges, around skyscrapers, and over large structures. This way, gusting, channeling, and cornering effects can be fully observed and assessed, as shown in this article.

The Seaport Simulation Setup

Step 1: CAD Creation and Import

This simulation starts with a CAD model representing the Stockholm Royal Seaport development project topography and buildings. This type of simulation does not require watertight, or seamless geometries in order for the simulation to be performed.

Step 2: Simulation Type Selection

Step 3: Select Zones of Interest

The zone of interest is then visualized in the form of a circular zone, while changes to its location and radius can be made.

In the “wind condition” section, the wind rose data is imported to input the correct wind inlet profile for each direction for the analysis.

The next step consists of defining the zones you would like to obtain output results from. One can select a pedestrian zone at a certain height (1.5m) above the terrain, depending on local conditions and the height you want to obtain results from; foot-level, bench height for people sitting, or head height, for example.

The remaining simulation settings such as the precision refinements and the real-time of the simulation are automated according to the setup, and the simulation computes the calculations in all directions (16 directions in this case).

Download our ‘Tips for Architecture, Engineering & Construction (AEC)’ white paper to learn how to optimize your designs!

Download Now

The Simulation Results

The results of each run were then accessed from a standard web browser, and any output quantities in the form of cutting planes, streamlines, isosurfaces, etc. were visualized in both a transient and average state.

The SimScale online post-processor integrates most of the wind comfort criteria in the industry such as:

• Lawson
• Davenport
• NEN 8100

These criteria essentially combine the calculation results from all of the wind directions and analyze the frequencies of the wind at certain thresholds of velocities. This takes into consideration the frequency at which each point the wind reaches velocities above 5m/s, and this, for all wind directions combined.

This powerful visualization and numerical assessment tool aims to give architects, urban development engineers, and all the stakeholders of projects like the Stockholm Royal Seaport plan insights and meaningful information in order for them to make the best design decisions to produce a sustainable, safe, and comfortable environment for all the residents and pedestrians.

The post Sustainable Wind Engineering: The Stockholm Royal Seaport Project appeared first on SimScale.