UpfrontCFD talked to Jason about SmithGroup, sustainability, and the engineering tools that are fueling the green architectural movement.  Excerpts of the interview are below.  The full interview, along with more than a dozen articles on CFD, simulation and sustainability can be found in the eZine UpfrontCFD: The Built Environment Issue, available for free download.

Jason Sambolt, LEED AP, SmithGroup

What specific design, engineering and construction techniques are helping you improve LEED ratings and meet the rigorous energy standards of the Architecture 2030 Challenge?

There are multiple design approaches SmithGroup has used to improve building performance over the years. 

Constitution Center in Washington, D.C., was the first large-scale chilled beam installation in the United States.  The active chilled beam system in combination with dedicated outdoor air units with energy recovery produced significant energy savings in cooling, heating and fan energy consumption. 

SmithGroup used CFdesign to determine whether a building that originally implemented natural ventilation in 1881 was capable of providing enough natural ventilation after 100 years of urban growth around it.

A natural ventilation design strategy was implemented at the Chesapeake Bay Foundation Headquarters which helped in achieving the first LEED New Construction Platinum rating.  

 How does upfront CFD fit into your sustainable design efforts?

As the demand for sustainable design increases, so does the demand to be on the forefront of new technologies and to increase outside-of-the-box thinking.  This often means dealing with long-term untested practices.

An atrium temperature study conducted by SmithGroup for the L’Enfant Plaza retail renovation.

Upfront CFD has allowed SmithGroup to confidently implement new technology and ideas through proof of concept verification.  A room with a chilled beam system can be modeled quickly and easily during the design process to ensure thermal comfort will be maintained.

Another good example is the use of CFD to determine the feasibility of a natural ventilation design for a retrofit of an existing building.  SmithGroup used CFdesign to determine whether a building that originally implemented natural ventilation in 1881 was capable of providing enough natural ventilation after 100 years of urban growth around it.

At what point do you bring CFD into your design/engineering work for a new building?

At SmithGroup, architectural and engineering design teams collaborate at very early stages of design in order to create a fully integrated practice.

During these early design stages important decisions need to be made based on schematic-level energy analysis and CFD studies that greatly affect the overall design.  A relatively small effort early can produce invaluable information to help both the architect and engineer better understand a project.  This information not only ensures we are creating an energy-efficient building but that we are also producing the best product for our client.

What kind of cultural changes are required to integrate CFD into design and engineering processes?

The biggest change needs to be early integration.  Often times consulting engineering firms will be brought onto a job after the early phases of design have already been developed by an architectural firm.  By this point critical decisions have already been made without a full understanding of the design.

Do you think sustainability is going to continue being a watchword for the future?

Sustainable design is here to stay.  As we progress, the challenges of creating more environmentally responsible designs will increase and will begin to push the envelope of building design and technologies. This challenge will require architects and engineers to constantly expand their perception of sustainable design and push current boundaries to reduce our overall impact on the environment.

###

Underfloor air distribution (UFAD) is becoming a popular alternative to traditional ventilation systems, improving air quality and thermal comfort while reducing energy consumption.

In the U.S., LEED and Energy Star are frequently specified by clients, and BREEAM, NABERS and CBA Green Mark, among others, are prevalent internationally.  Inextricably linked to these standards are digital technologies such as BIM and computational fluid dynamics (CFD) that emphasize sustainability very early in the design process.

A prime example of the impact of CFD can be seen in the LEED points system:  More than half of the possible points a building can be awarded go to areas where CFD adds considerable value, including energy and atmosphere, indoor environmental quality, and innovation in design.

At its core, sustainable building design is about ensuring that factors such as aesthetic appeal, occupant comfort, indoor air quality, and safety needs are considered for both the present and the future.  CFD informs energy-efficient design and takes advantage of natural resources to significantly improve sustainable development.

Understanding cause and effect

Upfront CFD – defined as using CFD early in the design and engineering process – provides real-world validation and optimization directly from building information modeling (BIM) systems.  Architects and engineers are able to understand how design changes can help achieve environmental objectives, meet government certifications, and increase human comfort over the long term, all well before anything is built.

Jonathan den Hartog, AEC application specialist with for Blue Ridge Numerics, cites the following ways upfront CFD can help maximize energy efficiency and maintain air quality:

• It enables performance of a new technology or design approach to be studied for effectiveness early in the development stage before significant time investment or design resources are committed.  An example: natural ventilation, such as fresh air cooling for datacenters.
• It gives architects and engineers a way to demonstrate how a design approach will achieve energy efficiency standards, energy certifications or air quality requirements before construction.
• It offers tools for maximizing efficiency of a given design approach, such as varying the thickness of the air layer in a ventilated façade.
• It provides tools for making fast, accurate comparisons of thermal comfort and air quality issues for different types of HVAC and structural designs.

CFD for mere mortals

What makes this all possible is the evolution of CFD from highly specialized software requiring years of expertise to software that can be used by designers, architects, engineers and others right out of the box.

“A cultural shift is underway, marked by a high rate of CFD adoption among AEC and MEP firms,” says Parker Wright, AEC segment manager for Blue Ridge Numerics.  “It’s driven by a combination of increasingly stringent sustainability requirements, continuing computer hardware advances, and CFD software that is now accessible to mere mortals.  Integrating CFD into the design process is helping firms increase agility, differentiate themselves from the competition, mitigate risk, and positively impact the bottom line.”

Architects and engineers sharing

Both architects and engineers can benefit from using CFD in early development stages, according to den Hartog. 

“Architects and engineers can leverage existing geometry and share a similar understanding of results and design impacts.  Architects may be involved in the very early stages, such as looking at building massing or façade material studies.  Engineers might examine performance of somewhat more detailed models and determine, for example, whether or not a ventilation concept is feasible or to characterize wind loads on a structure.”

Diversity of applications

The democratization of CFD goes beyond who can use it and when.  The newest generation of CFD software accommodates a wide range of architectural studies, from micro (diffuser) to macro (master planning).

Studying solar gains on facades allows early master planning decisions on building placement and orientation. When combined with wind analysis, firms can more creatively design spaces that are activated and enhanced by the sun and prevailing breezes.

“Thermal comfort, humidity, radiant and solar panels, wind/wake effects, smoke migration, exhaust concentrations, and stack re-entrainment are just some of the applications where CFD is valuable,” says Wright.  “It’s extremely effective for any application with flow or thermal design implications.”

Size doesn’t matter

Size – of the project or the firm – doesn’t matter with upfront CFD, according to den Hartog. 

“Upfront CFD can be applied to almost any architectural project, ranging from component-level studies to wind/wake studies on the scale of a city block.  Similarly, it can be leveraged by large and small firms.  Large firms typically have the capacity to have their engineering team apply CFD across a range of different applications.  Small-size firms might specialize in a particular application – such as datacenters or clean rooms – and use CFD as a competitive differentiator.” 

Wright says that all it takes is one motivated individual to start reaping the rewards of upfront CFD. 

“There is a large contingent of consulting companies with three employees or less who are tremendously successful with CFD.  Many have more work than they can accept right now because of the sales, marketing and engineering edges this technology provides.”

Large AEC and MEP firms are claiming similar success, using CFD to share results with clients, improve communications, create new green initiatives and compress development time.

Growing hand in hand

Given the compatibilities between sustainability goals and upfront CFD, it’s no surprise that leading vendors such as Blue Ridge Numerics grew by percentages in the upper teens during the past year, despite the weak global economy.

Demand is likely to increase as CFD becomes more widely recognized as a major vehicle for ensuring sustainability, whether defined as energy savings, air quality or human comfort.

As Jason Sambolt of SmithGroup concludes in the interview within the Built Environment issue of UpfrontCFD, the eZine:

“The challenges of creating more environmentally responsible designs will increase and will begin to push the envelope of building design and technologies.  Upfront CFD has allowed SmithGroup to confidently implement new technology and ideas.”

 ###

Read more about sustainability in the built environment in the 26-page eZine UpfrontCFD: The Built Environment Issue, now available for free downloading.

Visualization of turbulent airflow patterns within the existing structure at Yale School of Medicine.

“Typical guidelines say that more air means fewer contaminants lingering in a room,” says Marty Wallace, a senior associate at Genesys.  “But high ventilation rates cost large sums of money and are not always necessary to control contaminant levels.”

As with any new approach, Genesys had to prove its point.  That’s where upfront computational fluid dynamics (CFD) comes into play.  Genesys worked with engineers from Blue Ridge Numerics, who used their own CFdesign software to simulate a new design that promises not only to eliminate contaminants more effectively, but to do so with nearly half the airflow of the existing configuration.

 Aggressive plan

The project taken on by Genesys is part of an aggressive plan to cut the Yale School of Medicine’s energy consumption 25 to 30 percent below the criteria set forth in ASHRAE Standard 90.1, which provides guidelines for energy efficiency within the HVAC industry.

The first building to be tackled under the plan is The Anlyan Center.  It is the largest structure at the Yale School of Medicine, consuming 25 percent of the medical school’s total energy.  The eight-story building, built in 2001, comprises 475,000 square feet of research labs, an MRI suite, classrooms, offices, and support space.  Initial work by Genesys centers on improving efficiency for a 1,800-square-foot lab in The Anlyan Center that is typical in size and features. 

Combined expertise

Genesys and Blue Ridge Numerics were natural partners for this project.

Genesys has performed energy evaluations for more than 250 buildings over the past four years.  The company specializes in laboratory facilities in universities, hospitals, research institutions and industry. 

In addition to its CFdesign software, Blue Ridge Numerics brought to the table its experience in optimizing HVAC designs for computer-server rooms, retail stores, museums, office buildings, clean rooms, aircraft hangers and other diverse structures.  The company’s CFdesign software is categorized as an “upfront CFD” product.  It enables mechanical engineers to conduct fluid flow and heat transfer simulation early in the product development process, where it can save the most time and money.

“We could not have proven our point without the simulations provided by CFdesign software,” says Wallace.  “Upfront CFD was the only way we could analyze airflow patterns with any degree of confidence, without the exorbitant expense of constructing a full-scale model of the space.”

 ”No stone unturned”

Genesys started the project by studying the original drawings of the building, examining test and balance reports that document air quantities being delivered by the HVAC system, and reviewing control system schematics.  The research was supplemented by hours of interviews with the school’s operating staff and hundreds of hours on-site examining every aspect of the building.

“We had to determine how much systems could be modified given the constraints of existing construction,” says Wallace.  There was also some sensitivity to the fact that the building is only eight years old, with systems considered fairly new by some standards. 

“To its credit, Yale instructed us to ‘leave no stone unturned’ in our quest to find ways to save energy,” says Wallace.  “We could see that current flow patterns were not enabling effective ventilation.  The Yale School of Medicine staff agreed, and supported the idea of using upfront CFD to help illustrate the problem and proposed solutions.”

 Initial turbulence 

Parker Wright and Jason Pfeiffer of Blue Ridge Numerics kicked off the analysis part of the project by visiting The Anlyan Center and taking detailed measurements of the lab room, capturing major features such as pendant lighting fixtures and bookshelves that affect airflow.

Wright took the measurements and existing information on airflow rates and diffuser deflection angles and built a CAD model of the existing room in Autodesk Inventor.  The model, containing only the essential features that impact airflow and contamination removal, was meshed automatically by CFdesign to ready it for simulation. 

The simulations based on the existing layout showed exactly what Wallace thought they would.

“The CFdesign simulations proved beyond a reasonable doubt that the patterns were turbulent and not conducive to moving contaminants out of the room very effectively,” says Wallace.  “They also validated many of the occupants’ complaints about uncomfortable drafts in certain parts of the room.”

Visualizing the best solution 

Wallace provided Wright with a modified diffuser layout that would provide more uniform flow from the windows, down the bench aisles and to the fume hood and general exhausts.  Although an improvement over the existing layout, the revised design resulted in too many drafts in the room when simulated in CFdesign.  Two more diffuser locations needed to be designed and tested before Wallace and Wright settled on the best option.

Moving from CAD to CFD and back again was made quicker and easier by CAD integration within CFdesign.  Geometry from the CAD system was read directly into CFdesign and all CFdesign parameters were preserved when geometry was changed within the CAD system.

Wright modeled the best option in CAD.  He then brought the model into CFdesign for testing with several air change rates, simulating heating, peak cooling and emergency flush modes.  He also ran several chemical spill simulations to find out how long it took to clear most of the vapor from the room. 

“Diffuser placements and throws have to be designed in concert with hood flows from support equipment and exhaust outlets,” says Wright.  “Complex interaction among different elements couldn’t be fully understood until we ran the CFdesign simulations.

“The final arrangement promotes laminar (smooth) flow from diffuser (inlet) to exhaust grilles, clearing chemicals, reducing drafts and resulting in a more sustainable design,” says Wright.  “The results showed conclusively that the new layout, with less airflow, cleared the room of contaminants more quickly and effectively than the original layout.”

Post-design studies with modified ducts and diffusers proved that room temperature was even and hot spots eliminated.

Having proven the premise that more is not necessarily less – at least as it applies to airflow and contaminant removal – Genesys has completed a pilot project that modifies the lab room’s ducts and diffusers based on the results of CFdesign simulation. Early reviews of the changes are excellent, according to Wallace: Room temperature is even, hot spots are completely gone, and occupants have never been more pleased.

“Blue Ridge’s upfront CFD capabilities saved us time, allowed for more iterations, and showed us something we couldn’t have seen otherwise,” says Wallace.  “It’s tough to quantify, but I just can’t say enough how much easier it is to understand an airflow problem when you can see the graphic results.”

### 

Bob Cramblitt is principal of Cramblitt & Company in Cary, N.C.  He writes about technologies that dramatically impact the way products are designed, engineered and manufactured.

Figure 1 – “stair step” approximation for a curved surface

The FVM is a variation of the finite-difference method (FDM). Early CFD researchers used the FDM for work on brick-shaped structured meshes.  This structure provided well-defined grids that could be constructed to relate the nodes or vertexes of the volumes in the analysis domain.

From FDM to FVM

The FVM method considered the flux of the dependent variable from node to node instead of the mathematical expression used in the FDM. Since the fluid is flowing through the analysis domain, it was a natural and intuitive way to represent the physics of the fluid. With the FVM method, however, the fluxes must balance exactly in order to get a viable solution. The FVM transforms the analysis goal from being a solution at the nodes to a solution of balanced fluxes.

Early FVM algorithms relied on structured or brick-shaped analysis domains because the fluxes were more readily understood by developers and easier to implement in a CFD computer program. Some FVM CFD codes still rely on this structured mesh analysis domain.

But, the structured mesh has some problems if you’re modeling anything but bricks.  With any type of organic or complex shape, you’ll not get a very accurate representation of your analysis geometry.

Stair steps and compute resources

Figure 1 shows the structured mesh representation of a curved surface. You can see the classic “stair-step” effect. As you can imagine, there are inherent inaccuracies associated with treating curved walls with this stair-step shape.

Another issue with the structured mesh is the amount of compute resources needed to carry a finely meshed area all the way through the model. Figure 2 shows the fine mesh that is required to model a two-dimensional cascade. Note that the fine mesh built around the foil shape has to extend to the boundaries. For three-dimensional bodies, the structured mesh must maintain the same number of elements in each of the three coordinate directions; the fine mesh must extend far beyond the region of high gradients.

Figure 2. Structured mesh on a two-dimensional cascade (http://www.iafr.eu/TESI/4.htm)

So, why don’t the FVM developers use an unstructured mesh?

Some do, but it requires a massive effort to convert the FVM to a completely unstructured mesh.  Even then, numerical instabilities abound due to the nature of unstructured meshes. Compute resources are also taxed because of the tremendous amount of bookkeeping required to keep track of which fluxes should be attached to which nodes.

The FEM for CFD

In the early 1980s, CFD researchers began exploring the finite element method (FEM). Most of the early adopters were already having success using finite elements for structural analysis. In fact, the FEM has been the standard for structural analysis (FEA) for decades.

Extending FEA to the CFD world wasn’t easy, however. Early adopters of the FEM found it difficult to apply the lessons learned from the FVM work already done. By this time, the algorithms developed for the FVM were well-honed to be efficient for solving difficult CFD problems, but they resulted in inefficient meshes. Because the governing equations for CFD are coupled and non-linear, approximations were necessary.  Early FEM researchers were not familiar with the approximation methods or found them too “un-mathematical.”

I was part of an early group in the mid-1990s at the University of Virginia that had a great deal of industrial experience developing FVM codes for the electric power industry. We decided to use the basic-weighted integral approach of the FEM to discretize the partial differential equations (PDEs) for fluid velocities and pressure.  We combined this approach with the overall solution algorithms and other numerical approximations (such as upwinding) of the FVM.

The combination of these techniques resulted in a CFD code that incorporated the most computing-efficient aspects of the FVM, but could more accurately, robustly and efficiently model the difficult geometries of the real world.  While the solution incorporates both methods, the basic foundation is the FEM, so we called it FEM-based CFD.

Best of both worlds, but mostly the FEM

FEM-based CFD was successfully demonstrated on some of the most difficult CFD problems, including turbomachinery, combustion, high-speed external flow and many more. It became the basis for the code used in CFdesign software.

From a CFD user’s perspective, there are critical advantages in the FEM-based approach:

• CFD solutions can be more easily obtained, since you are solving a FEM-like minimization problem.
• Meshing complex geometries is easier and more accurate with the tetrahedral meshes common to the FEM.
• Linking up CAD and other CAE analyses (such as FEA) is more readily accomplished with the FEM.

The principal advantage of the FVM is the more advanced physics that have been developed for this method due to its earlier adoption by CFD developers.  This advantage, however, is diminishing in conjunction with the rapid progress of new and improved finite-element methods within the CFD community.

With their unique background of developing CFD code using both the FVM and FEM, CFdesign developers can pull the best from both worlds, but the core methodology is still solidly FEM.

###

Rita Schnipke, PhD, has been a CFD software innovator for more than 30 years. She co-founded Blue Ridge Numerics Inc. with the express purpose of delivering a CFD tool suitable for mainstream product development processes.

Do you agree or disagree with this article? What do you have to add to the FVM vs. FEM debate?  Comment here.

How do you handle CAD integration?

From our perspective, native, associative CAD integration is the first tenet of upfront CFD. All geometry changes must be made in the MCAD system and the CFD tool must automatically and instantaneously recognize and adapt to design changes without loss of product data. CFdesign maintains this level of integration with SolidWorks, Autodesk Inventor and Revit, Pro/Engineer, Catia, UGS NX, Solid Edge, CoCreate and SpaceClaim.

Our approach eliminates the need for importing or translating a product design into the analysis environment, a time-consuming process that induces error and strips the geometry of all intelligence (part IDs, material properties, etc.) assigned in the MCAD environment.  It also disconnects the assembly from essential product data such as the BOM, assembly constraints, tool paths and drawings. 

I find it ironic that most CAE companies are pushing “simulation-driven design” without realizing the only way to deliver on the promise is to allow the MCAD model to drive the simulation process. When MCAD drives CAE then you can have simulation-driven design. 

Why not form partnerships with CAD vendors to deliver embedded CFD?

Our approach to integration requires close partnerships with MCAD developers, but there are a few important reasons we don’t fully embed CFdesign into the MCAD environment.

First, we view it as inefficient use of resources. The time and manpower needed to maintain embedded applications from release to release for nine MCAD systems is huge. This approach would diminish our ability to deliver new analysis capabilities essential to the mechanical design market. 

Second, embedded applications tie up the MCAD seat during the meshing and analysis processes. This is a big penalty to the customer. 

Finally, if the MCAD integration is done correctly and the CFD interface makes it simple enough to set up and launch an analysis, the user really doesn’t care if it’s embedded.

###

What’s the best interface between MCAD and CFD? Contribute your comments.

I find the Harvard Business Review to be a great resource. It expands my horizons and makes me look like one of the smarter guys in the coach section of the airplane. Yeah I might be stuck in a middle seat of row 44 but look at me… I’m reading HBR.

Every month I eagerly crack open my new issue and start digging through for ideas that can make me and my company more successful. Sometimes this can be a stretch – since the editors rarely ever shine a spotlight on CAE, CAD, or even  product engineering.

PLM for the HBR crowd?

So imagine my surprise when I stumbled upon a colorful, full-page advertisement for Product Lifecycle Management software! Why would a PLM company spend their precious marketing dollars on the HBR crowd? Clearly they’re on a “whale hunt”. They’re hoping to beat the odds and actually develop brand awareness with a CEO or CFO of a Fortune 500 company and – in doing so – grease the wheels for one of their fortunate sales people.

More intriguing to me is that the advertiser used CFD as the heart of their value proposition, complete with cool visual from an aerodynamic simulation. Think about it… you’ve got the attention of a powerful executive and only 10 seconds to help them see the strategic importance of PLM to their global business. What would you say? Apparently… according to some folks who I’m sure devoted a ton of time wrestling with this question… the best angle is Computational Fluid Dynamics.

A little whale-speak

Of all the functionality and benefits contained under the PLM umbrella – CFD was the most compelling and easiest to understand. If you’re an engineer and you’re thinking your CEO “doesn’t get” the value of simulation tools like CFD maybe it’s time to re-think. Maybe it’s time to take to a page from the PLM boys and go whale-hunting. As the guys writing the big checks in your company emerge from their Great Recession bunkers they will be asking “where should we be investing to grow our business”.  

Show your boss how to take Engineering into the Digital Age, eliminate a ton of unnecessary physical testing, and make your company a poster child for quality and innovation. Most of them will “get” CFD if you’re willing to use a little whale-speak.

###

Jim Spann is vice president of marketing for Blue Ridge Numerics. He has nearly 20 years of experience in marketing and sales of CAE products.

 

Title	Votes	Rating	Review
An expert look at FE meshing	12	8.7	-1.0
CFD software debate: FVM or FEM?	28	3.4	-1.0
CFD: Integrated or embedded with CAD?	4	3.3	-1.0

A simulation generated by CFdesign software shows a plume of particulate coming off the iron tilting runner into an awaiting bottle car.

When Severstal North America recently rebuilt one of its blast furnaces, we faced a challenge: the emission control system had to be designed for a casthouse that didn’t yet exist.  The project would have been difficult – maybe nearly impossible – without upfront computational fluid design (CFD) software.

The redesigned casthouse included a new set of hoods, ductwork, fabric filtration equipment and fans.  Under U.S. Environmental Protection Agency (EPA) regulations, casthouse particulate emissions must be controlled with what is called Maximum Achievable Control Technology (MACT).  MACT compliance is measured by stack emission concentrations and the opacity of emissions that escape from the casthouse building.  The Michigan Department of Environmental Quality (MDEQ) also has established an opacity limit.  In our case, the casthouse needed to meet U.S. and Michigan environmental regulations by capturing 98 percent of emissions.

MACT compliance is usually based on actual observations of casthouse operations to determine emission intensities, crosswind speeds and vertical rise rates.  But, Severstal could not do that because the rebuilt casthouse was very different in configuration and operation from previous ones.

Our solution was to use CFdesign software from Blue Ridge Numerics to conduct a comprehensive design study that simulated the new equipment under every conceivable operating condition.

Setting up the model

The first step was setting up the CAD geometry in CFdesign to include the physical boundaries and internal blockages to fluid flow within the casthouse.  The geometry of the model, in this case created in MicroStation software, included the physical bounds of the casthouse as well as the furnace, tuyere (the outlet through which air is blown into the furnace) platform, iron and slag runners, hoods and other equipment.

After setting up the geometry in MicroStation, the model was meshed in CFdesign, which provides built-in tools for simulating the conditions that the new equipment and structure would face.  For the casthouse model, we needed to simulate the temperature and velocity of incoming air; temperature, velocity, fume and combustion components associated with the casting and pouring of hot metal; and air flow rates for the tilting runner duct, taphole hoods and slag pot shanty duct.

Information describing these processes must be incorporated into the model as boundary conditions to accurately evaluate the operation of the fume hoods that control emissions.  In this case, input information was derived from previous experience at similar casthouse operations, where heat and material balance calculations were coupled with video analyses.

 Establishing the parameters

After the model geometry and simulation information were entered, the initial hood designs were incorporated into the model to evaluate their performance.

Once again, heat and material balance calculations and video of similar casthouses were analyzed to establish fume rise velocities, intensities and horizontal movement, especially at the taphole, which is the most concentrated source for emissions.  From these previous studies, the following parameters were established:

• metal temperature – 2800° F
• slag temperature – 2700° F
• rise rate/vertical velocity – 900 ft./min. at the taphole
• cross drafts – 450 ft./min. at the taphole (approximately 5 mph) and 900 ft./min. under tilters

Other values that were assigned based on previous studies included the correspondence between fume intensity and emissions, and the fume densities for the taphole, tilting runner and slag pot shanty.

 Simulating real-world conditions

After establishing the fundamental geometry and base set of conditions, finding the optimal design was an iterative four-step process:

1. Using initial hood designs and ventilation rates, determine capture efficiencies for tapholes, iron tilters and slag pot areas.
2. If initial hood/ventilation combinations do not achieve the 98 percent emission control rate, revise the geometry of the hood and/or the ventilation rate.
3. Re-run the model with the revised hood/ventilation design.
4. Repeat steps 2 and 3 until acceptable emission control is achieved, then apply the acceptable model parameters to the emission control system design.

Each time a simulation was run, CFdesign performed a series of complex calculations and then compared them to determine the degree to which they agreed.  The software then automatically adjusted the model and reran the simulation to achieve close convergence for small sub-sections in terms of temperature, velocities and other parameters.  It required 225 to 250 iterations of each model to achieve convergence and calculate a valid capture efficiency.

Simulation of a plume of particulate coming off the slag tilting runner into an awaiting ladle.

Capture efficiency for the casthouse model was computed as the probability that a fume particle generated at each source (taphole, tilter or slag shanty) would be captured by each hood and subsequently drawn through the ductwork to the bag house.  The next – and most important – step in the calculation was to determine fume capture percentages for each individual operation.

The entire study entailed dozens of CFD simulations to arrive at an optimal mix of hood configurations and ventilation volumes.  A summary of the key results from CFdesign is shown in the table accompanying this article.  In addition to numerical results, CFdesign provided static and dynamic images that created a better understanding of what was occurring during the simulations (see accompanying images).

Summary of key CFdesign findings

Accurate and cost-effective

Upfront CFD using CFdesign software proved to be an indispensable analytical method for optimizing design of the emission control system for the Severstal casthouse.  It enabled us to successfully establish ventilation volumes, hood configurations and volume distribution profiles for 50 separate operating scenarios for the new blast furnace.  All this was done in three months, without costly physical testing.

At this time, the new furnace is fully operational and in compliance with the MACT regulations. Upfront CFD proved to be an accurate and cost-effective method to predict actual performance of the casthouse emission control system under real-world conditions.

 ###

A hexahedron (six-sided) element

The partial differential equations (PDEs) that we solved in our undergraduate education don’t even scratch the surface of real applications. It turns out that the PDEs describing fluid motion, heat transfer and lots of other stuff are impossible to solve even for Einstein, Newton or (insert your favorite genius).

So what do we do? Approximate, of course!

Getting down to element level

It turns out that you can approximate any PDE with a much simpler algebraic expression. The approximation is, of course, not very good; unless you apply the approximation in a small enough space. Then, the approximation does a pretty good job of mimicking the real physical behavior of the PDE.

The trick is to reduce your PDE analysis problem to a bunch of inter-connected smaller spaces – lots of smaller spaces (say 100,000-1,000,000 or maybe more, sometimes much more). Hence, we have to break the problem up into a mesh or grid of tiny volumes or elements to apply our approximate algebraic expression of the PDE.

An element can theoretically be any polygonal shape: four-sided like a tetrahedron, five-sided like a pyramid, six-sided like a hexahedron, etc. Because of its simplicity, it is easiest to fill arbitrary geometric domains with four-sided tetrahedral.

Defining the algebraic expression

How does the PDE shrink to an algebraic expression by the use of an element? Let’s go back to the approximation. On an element, we assume that the solution – temperature, velocity, pressure, etc. – takes a certain shape or algebraic expression. The shape the solution takes is dependent on the element shape. The number of terms in the algebraic expression corresponds to the number of vertices on the element shape. On a tetrahedron, for example, the solution will be an algebraic expression with four terms for the four vertices on a tetrahedron.

Let’s take a closer look at the concept of algebraic expression.  Assume that we have a PDE with a solution F(x) shown as the black line in the figure below. Let’s approximate that solution by a straight line with a very easy algebraic expression. Then the solution would look like the red line in the figure below. Not a very good fit to our solution, is it?

But, suppose we use two straight lines like the blue lines in the figure below. That’s better, right? Even better, if we use three straight lines like the magenta ones. So, you can see that as we break up the solution into smaller segments, we are getting closer to the more complex shape of the solution.

This is exactly what we do with the finite element mesh for fluid flow and heat transfer analyses.  One other thing that you can see from the figure below is that you can use a larger-size element or breakup at the left end of the F(x) expression than you need on the right end of that function. This is also true in PDE problems like fluid flow and heat transfer. There will be areas with large solution changes that require a smaller mesh and other areas that can be done with a larger mesh.

Now that we know what the algebraic expression is, we can substitute that into the PDE. In CFdesign, we integrate the PDE over that element volume to arrive at the solution for that element. But, the element solution doesn’t help us much because we want to know the solution everywhere. So what do we do?

Interconnected nodes

We gather the element equations at the vertices or nodes of the elements. The elements that form the geometry are all interconnected at the vertices or nodes of the individual elements. When we gather the information on each element to the nodes of the elements, we arrive at a giant matrix solution at all of the nodes on all of the elements, hence the solution in the entire geometric domain. This solution at the nodes is what you see plotted in the graphics window of CFdesign.

Sounds easy enough, but it is actually pretty hard to fill the problem geometry with all those elements. Imagine trying to fill the room that you are sitting in with millions of tetrahedral elements that all have to be connected. Every tetrahedral element face has to touch another tetrahedral element face unless it is on the boundary of the geometry.

There are many ways to mathematically do this filling job, but none of them are fool-proof. If you thought fluids or dynamics was hard, meshing is at another level of difficulty. In addition, you cannot just fill the space in any old fashion. The best mesh is one that has tetrahedrals with the same size element faces and angles of 60^o. It would be impossible to fill anything but a cube with elements like that.

Flatten and stretch

The solution is to flatten and stretch elements to exactly fit the space. If we make the elements too flat (almost planar) or too stretched (one node far away from the other three – high aspect ratio elements), the approximation and integration gets really ugly. Translated, that means the solution is hard to obtain (lots of iterations and maybe the solution will not converge) and/or the solution is not highly accurate.

Stretched element Flattened element

Flattened element

In CFdesign, we have implemented several solver techniques to work around bad meshes. Ninety percent of the time, CFdesign will get a usable solution if the geometry meshes (available for upfront, quick analyses). Of course, you can also refine and adjust the mesh and setup to hone the accuracy of the solution to within a few percent.

Meshing headache medication

The automatic mesher in CFdesign removes a lot of the headache of making good meshes. We intelligently examine the curvature, size and proximity of the geometric features of the CAD geometry to estimate an element size. Then using empirically derived growth algorithms we fill the volume starting with this size, stretching and growing where appropriate, and always controlling the placement of the elements so that their size will vary slowly.

Boundary layer meshing in CFdesign automatically creates a skin of elements around each geometric feature to best capture the physics at those locations, allowing you to gauge the effectiveness of the design.

All of this automatic meshing technology is based on our own decades of experience as well as thousands of CFdesign users world-wide. We are constantly tweaking and adjusting these algorithms as we daily learn more about best practices in meshing.

###

Rita Schnipke, PhD, has been a CFD software innovator for more than 30 years. She co-founded Blue Ridge Numerics Inc. with the express purpose of delivering a CFD tool suitable for mainstream product development processes.

Brookings is the home of Daktronics, one of the world’s largest suppliers of electronic scoreboards, computer-programmable displays, digital billboards, and large-screen video displays and control systems. 

Out of its 500,000 square-foot manufacturing and office facilities come the displays that show sports fans scores and video replays, advertise new offerings for all types of businesses, tout the stars appearing in Vegas nightspots, and help drivers navigate highway systems.  In the case of the Grand Lisboa Hotel and Casino in Macau, China, a Daktronics display is a spectacular calling card that heralds the excitement to be found inside.

 Complexity behind the lights

Daktronics products require integration of complex multiple displays showing real-time information, graphics, animation and video.  They have to be built to withstand all types of weather, to dissipate heat generated by LEDs, and look good even when directly opposite an unforgiving sun.

The tool that helps Daktronics analyze these complex factors early in the design process is CFdesign upfront CFD software from Blue Ridge Numerics.  CFdesign software has become an integral part of Daktronics’ product development workflow, giving the company a picture of how designs will perform before major commitments of time and resources are made.

“We use CFdesign in order to thoroughly understand complex electronics cooling situations and make comparisons among different designs before we start to build,” says Shannon Mutschelknaus, thermal product development engineer at Daktronics.

A natural convection, solar and electrical heating analysis generated by CFdesign software for a section of a Daktronics display system.

 Benefits vs. cost upfront

A typical project at Daktronics starts with a Pro/ENGINEER model of the display.  Native geometry from Pro/E is used by CFdesign to create the analysis model, eliminating the time-consuming translation process required for traditional CFD.

 “The CAD/CFD integration is very important to us,” says Mutschelknaus.  “We want to spend our time exploring design options, not preparing files for CFD.”

 CFdesign is opened directly in Pro/E using controls within the CAD software’s interface.  Flow volume, volumetric boundary conditions, and material properties are assigned automatically.  After that, all that is left is selecting flow and heat-transfer analysis options.  CFdesign automatically generates the optimal mesh and provides access to initial simulation results within minutes. 

 For less demanding designs, Daktronics runs an airflow/ventilation analysis to compare different design options and optimize the airflow through the system.  For designs incorporating complex thermal cooling systems, Mutschelknaus and his colleagues, Sunil Gaddam and Kurt Peters, run natural convection and airflow analyses.  Associative data between CFdesign and Pro/E makes it easy to run a simulation, do a 3D design review, make appropriate design changes, and see the impact of those changes in minutes.

“We use CFdesign in the early stages of the design to compare benefits vs. cost for different combinations of components, fans, heatsinks, enclosures and materials,” says Mutschelknaus.  “With this information, we can narrow our options to two or three different designs that we’ll physically prototype.”

New solar loading functionality in CFdesign has also played a key role in Daktronics’ design work.  Solar loading depicts radiation through transparent media and even shows shadowing based on the sun’s movement.  Set-up is simple: specify the time of year, time of day, and location on the globe using a database within CFdesign or by assigning specific latitude/longitude coordinates, and click the mouse to see the simulation.

“Time-stepped solar simulation has enabled us to optimize display contrast by varying sizes and shapes of shading louvers on the display face,” says Mutschelknaus.  “It’s another way that upfront CFD has helped us determine performance in the early design stages.”    

Widespread impact

According to Mutschelknaus, nearly every project at Daktronics benefits from moving CFD upfront in the design process. 

“We’ve been able to make key improvements such as fan hoods with lower pressure losses, more efficient combinations of fans, tighter pixel pitches for outdoor use, improved display contrast, and reduced operating temperatures of electrical components.  We’ve even been able to design some displays without ventilation fans.”

A Daktronics display at Heinz Field, home of the Pittsburgh Steelers football team.

Although Mutschelknaus acknowledges that upfront CFD has probably saved Daktronics substantial time and money, his focus is on the greater opportunities it has opened up.

“It has helped us engineer higher-quality products that are superior to our competition, break down existing design barriers, and define realistic expectations of a product earlier in the design process.”

At its core, breaking down barriers and defining realistic expectations are what good design is all about.  At Daktronics, that translates into everything from the utilitarian football scoreboard at a proud Texas high school to the spectacular light show adorning the facade of the Grand Lisboa in China. 

###

Bob Cramblitt is a technology writer who focuses on new developments and processes that make a definitive difference in how we work and live.