Leslie Langnau

Avoiding singularities in FEA boundary conditions

February 26, 2021 By Leslie Langnau Leave a Comment

FEA boundaries can usually be obtained using simple fixed constraints. But, this can result in singularities that produce erroneous results. To determine whether results show a real stress concentration or a singularity, an accurate solution can usually be obtained by either using elastic supports or modeling contact between components.

Singularities caused by sudden changes in boundary conditions can be harder to spot and resolve. In fact, setting up realistic boundary conditions is often the most challenging aspect of a simulation.

Dr. Jody Muelaner, PhD CEng MIMechE

Singularities in Finite Element Analysis (FEA) can cause real issues, even for an apparently simple structural analysis. Singularities lead to completely erroneous results and stresses that continue to rise as a mesh is refined. Many singularities are caused by stress-raising geometry such as holes and sharp internal corners, and this is generally well understood. Singularities caused by sudden changes in boundary conditions can be harder to spot and resolve. In fact, setting up realistic boundary conditions is often the most challenging aspect of a simulation.

What is a singularity?
A singularity is a point in the model where a value, such as stress, tends to infinity. As the mesh is refined, the increasingly small elements get closer to this point and the value therefore rises. As the element size tends to zero, the stress will tend to infinity. This produces nonsensical results and prevents mesh convergence.

Geometry that causes singularities
Singularities caused by stress-raising geometry such as holes and sharp internal corners are well understood. In the real world, there is likely to be a small radius on any internal corner, meaning the stress would not actually continue to rise. In any case, local yielding will limit the stress in such features. The location of these singularities can often be readily identified, excluded from convergence results and localized models used to predict the true stress in the features responsible. Singularities at corners are similar to cracks and the stress intensity factor can be calculated using the J-Integral, or considering the strain energy release rate – the energy dissipated during fracture.

There is, however, another type of stress raiser in FEA models that is talked about less often and which can be more difficult to deal with. Where there are abrupt changes in boundary conditions, such as a split line where a fixed constraint ends, this can also result in stress that continues to rise unrealistically and causes mesh convergence to fail. Let’s explore why this happens and how it can be avoided.

How can boundary conditions cause singularities?
The most obvious way that a boundary condition can cause a singularity is when a force is applied to a single node. Since stress is force divided by area, applying a force at a single point will give an infinite stress. If the area where the load is applied is not of interest, then it can be acceptable to use such a boundary condition. Due to Saint-Venant’s-principle, which states that, if the distance from the load is large enough, two different but statically equivalent loads create essentially the same effect. This can be easily seen where the same total force is applied to the sponges in two different ways. The fingers represent point loads and the flat hands distributed loads. Although the effects close to the applied loads are different, in the center of the stack, at a sufficient distance from the loads, the effect is virtually the same.

Saint-Venant’s principle states that, if the distance from the load is large enough, two different but statically equivalent loads create essentially the same effect. This image illustrates the point. The fingers represent point loads and the flat hands distributed loads. Although the effects close to the applied loads are different, in the center of the stack, at a sufficient distance from the loads, the effect is virtually the same.

When loads are simplified to point or edge loads, it is simply important to understand that the very high stress around the applied force does not represent reality. These regions must not be included in results, mesh convergence or adaptive meshing. Next, we’ll look at some more involved examples of boundary conditions casing singularities.

Abrupt changes to a fixed boundary condition
It is often convenient to fix a face of a model, to constrain a component that is loaded with forces applied in some other area. It should first be noted that such a fixed constraint can never truly represent reality. A fixed boundary condition essentially means that the face is bonded to an infinitely stiff body. In the real world, all solid bodies have some flexibility and often a part will actually be clamped rather than bonded. However, if the peak stresses are not expected in the region being represented by a fixed boundary, this may seem like a reasonable approximation. As with point loads, it can, therefore, be a good idea to simply exclude the stresses in this region from any mesh convergence. However, not all software allows this, and it is particularly problematic if automatic mesh refinement, or adaptive meshing, is being used.

A shaft can provide a good example of these issues. The shaft illustrated below has been cut in half and a symmetry fixture applied. The smaller cylindrical face at the right-hand end of the shaft has been split into three separate faces to allow a vertical force to be applied to a defined region. The other end of the shaft would be held by two bearings, with the outer bearing constraining axial movement against a shoulder and the end face.

The shaft has been cut in half and a symmetry fixture applied. The smaller cylindrical face at the right-hand end of the shaft has been split into three separate faces to allow a vertical force to be applied to a defined region. The other end of the shaft would be held by two bearings, with the outer bearing constraining axial movement against a shoulder and the end face.

When standard inelastic fixtures are used, stress singularities occur where the fixtures end. This effect is equivalent to the edge of a stiff part digging into a soft part. This can be seen below, where a bearing support extends between a split line and an external radius. It also occurs on the face of the shoulder which was constrained using a roller/slider constraint in SolidWorks Simulation. When the mesh is refined on the radius it is clear the singularity occurs where the fixtures end and is not a real stress concentration in the radius. This is particularly problematic because the radius is also a stress concentration and this region cannot, therefore, simply be excluded from the results or mesh refinement.

When standard inelastic fixtures are used, stress singularities occur where the fixtures end, as shown here where a bearing support extends between a split line and an external radius. This effect is equivalent to the edge of a stiff part digging into a soft part.

Using elastic supports
One solution is to use elastic supports rather than fixed constraints. In a fixed constraint, each node on the constrained surface is forced to zero displacement. An elastic support consists of an additional spring element for each node on the constrained surface. One end of the spring is attached to the node on the surface and the other end of the spring is fixed with zero displacement. The actual stresses in the springs are not normally included in the results. Using elastic supports can eliminate the issues with stress singularities at the edges of boundary conditions, but care must be taken to select realistic stiffnesses for the supports. If the stiffness is too small, the model may encounter excessive displacements which cause the solver to fail. On the other hand, if the stiffness is too great, a spurious stress concentration may still be seen at the edge of the constraint. Although an initial value for the support stiffness may be calculated by considering the type and thickness of the actual material which would provide the support. The image below shows that, with correctly set elastic supports, the model properly converges on the actual high-stress region.

With correctly set elastic supports, the model properly converges on the actual high stress region.

Modeling contact
Another similar approach is to model contact between the supporting components and the component of interest. This can be the most accurate, but can also be seen as simply pushing the problem to another area, since the supporting components must then be constrained in some way. However, if the supporting components can be excluded from the mesh convergence and final results, the supporting components can be simply constrained by fixing faces.

Mesh convergence and adaptive meshing
Mesh convergence is one of the most important methods to ensure a reliable FEA simulation. The basic process is simply to rerun the simulation a number of times, refining the mesh around areas of interest and recording the relevant values for the simulation using each mesh. When the value of interest varies randomly, and by a small amount, in both directions, the model can be said to have converged. If there are large differences in the result, or the result keeps creeping in the same direction as the mesh is refined, then this indicates a problem, often a singularity. What constitutes a small change is somewhat subjective but can generally be considered as a few percent of the value under consideration.

When the value of interest varies randomly, and by a small amount, in both directions, the model can be said to have converged. If there are large differences in the result, or the result keeps creeping in the same direction as the mesh is refined, then this indicates a problem, often a singularity.

Adaptive meshing takes mesh convergence a stage further, automatically refining the mesh at areas of interest and rerunning the simulation until the model is converged, or convergence fails according to some criteria. There are two types of adaptive meshing: H-Adaptive reduces the element size and P-Adaptive increases the element order. SolidWorks Simulation does not currently support P-Adaptive meshing for elastic supports.

Conclusions
In many cases, it may be possible to obtain useful results while using simple fixed constraints. However, this can result in singularities that may prevent mesh convergence and produce erroneous results. It is important to be aware of this issue, since some judgment may be required to determine whether results show a real stress concentration or a singularity arising due to simplified boundary conditions. When this happens, an accurate solution can usually be obtained by either using elastic supports or modeling contact between components.

A view on where AR/VR is headed, roundtable discussion from those who know

February 12, 2021 By Leslie Langnau Leave a Comment

Recently, Ron Fritz, CEO of Tech Soft 3D, hosted a roundtable discussion with five other industry executives to discuss the current state of augmented reality (AR) and virtual reality (VR). The core question at hand: whether AR/VR is finally poised for its breakthrough moment – and if so, what barriers might need to be removed to usher in this new era.

The participants included:

– Asif Rana, COO of Hexagon, a provider of sensor, software, and autonomous solutions

– Martin Herdina, CEO of Wikitude, an augmented reality technology company

– Susanna Holt, VP Forge Platform, Autodesk, a provider of 3D design and engineering software

– Thomas Schuler, CEO of Halocline, a developer of VR products for production planning and manufacturing

– Tony Fernandez, CEO of UEGroup, a user experience agency

A lightly edited and condensed version of the conversation and their unique perspectives follows.

Q: At various points over the past decade, many of us have believed that AR/VR was ready to really take off in the industrial setting – but it hasn’t happened yet. What are the barriers that are standing in the way of that widespread adoption, and what should the industry be focusing on?

Asif: One of the fundamental things that we tend to forget when we think about commercializing a technology is the user experience. I think one of the main hurdles of AR/VR in the commercial usage is we don’t think about the full user journey or what the full end-to-end solution looks like.

Martin: For a while, there was such a focus on technical benchmarks that nobody really talked about what could be achieved with AR/VR. Even when people did start to talk about what could be achieved, they didn’t really look at the full picture and at how things could be scaled beyond a single isolated use case. As long as that underlying basis is missing, widespread adoption of AR/VR will be hampered.

Susanna: I think one thing that’s lacking around AR/VR is pre-processing of data and data preparation – from CAD design data, to mesh poly count reduction. That kind of stuff needs to be automated, robust, fast, and scalable. And at the moment, all of that still seems to require too much manual work to really enable this AR/VR takeoff that we’ve been anticipating for the past 20 years.

Tony: I think the core issue is that AR/VR did not emerge from a human-centered point of view. It emerged from a technological exploration point of view. And what that has meant is that the human factors of this technology are terrible.

To take the case of VR: Who thought it was going to be a great idea to duct tape a TV to your head and blindfold you? Meanwhile, with AR, one of the problems that we continually run into is arm and body fatigue from having to hold up a device. Because AR/VR technology hasn’t centered around the reality of the human body, how it gets fatigued, and how people feel motivated to use their bodies, it will continue to have a difficult time breaking through to the mainstream, regardless of the value proposition it may offer.

Q: From what everyone’s saying, it seems that the user experience is one of the big barriers to mainstream adoption. What needs to be different for people to feel comfortable? How can companies remove this barrier?

Tony: I think mobile AR is a really difficult problem to solve. And again, part of the problem with most existing AR solutions is that they require people to use their bodies in unnatural ways. From a hardware perspective, we’re going to be much closer to solving that problem once we get to some sort of compact glasses. Of course, glasses come with their own problems around power and where to place the battery and so on. But I think that’s what AR’s waiting for, in terms of a hardware platform solution.

Asif: I wonder whether there are the same expectations on an enterprise level as at a consumer level for AR/VR. I say that because in the enterprise, you do see technology that’s not so comfortable to use – but it delivers such a high value that it’s used anyways. So, perhaps the AR/VR hardware is “good enough,” and it’s the content side that deserves more focus to deliver applications that can really make an impact and deliver value. Either way, I’d say that if the hardware companies focused on more business cases, that would be helpful to the enterprise sector.

Susanna: It’s true that the enterprise use case may put up with all sorts of inconveniences. But when I think of a use case for us at Autodesk, which might be an architect or structural engineer at a construction site or building site, inconvenience can quickly become a safety concern. AR provides a limited field of vision. In normal life, we don’t just look straight ahead – we’re constantly taking in things occurring on the periphery. Excluding that visual information in a potentially dangerous environment like a construction site does strike me as a risk factor. So, the hardware has to be natural to the way we conduct ourselves as humans in a particular environment.

Martin: I think the most important point that people have hit on is that things have to feel natural. When you wear a HoloLens, it’s cool, but it’s nothing that you would want to wear for 10 hours per day at your workspace. Another aspect that companies should address is the fact that so many AR use cases totally lack context. For example, why would you use AR to project a team roster on your desk when there are so many other user interfaces that make so much more sense for that objective? AR needs to really link reality to a reasonable set of content.

Q: Lots of big names – including Google, Apple, Facebook, and Microsoft, to name a few – are heavily investing in the belief that the barriers around AR/VR adoption are being resolved and that this an area that is ripe for explosion. All of your companies are, to varying degrees, investing in that belief as well. What makes you optimistic that AR/VR is getting close to a real breakthrough? What drives your confidence?

Thomas: It takes a long time to bring hardware technology from an early prototype to a usable product. You have to really keep at it for quite some time. What makes me optimistic is that the hardware vendors are still investing in it and pushing it forward – they’re not standing still.

At the same time, more and more content is now being produced that makes more sense. I think more people understand now that you need a different set of tools for AR or VR rather than taking the same old tools that you had before, but just manipulating them differently. So, while the progress might be slower than everyone expected, that progress is very much ongoing. That makes me optimistic that we are on an eventual path towards more widespread adoption for AR/VR.

Susanna: Well, let me turn this question the other way around. We’re hearing so much from our customers about how AR or VR is needed and how they’re expecting it to play a bigger role in their workflows. Some of that, of course, is a reflection of hype that they see in the media, but a significant proportion of it is a reflection of real need.

For example, while wearing a HoloLens headset might be uncomfortable today, it does allow you to make those important decisions much faster than having to look at something, take a photograph, go back to the office, think it through, discuss it, and so on. It will speed everything up. It’s about faster decisions, better decisions. There’s a real need in the market – so that bodes quite well for AR/VR, because a lot of technological advancement and evolution is driven by market need.

Tony: I would say AR/VR will break through if it can focus on its fundamental promise, which is to reveal information and perspectives in ways that would be difficult to do any other way. I’m not necessarily a believer that the way most companies have defined AR at this point is necessarily the path forward. For example, AR doesn’t necessarily always have to be visual in nature, right? It can be haptic in nature. It can be lots of other things. But visual is the primary road for now, and I think the need to visualize information that is otherwise difficult to do any other way or get access to any other way is going to drive the solution.

Martin: At my company, we perhaps have a unique perspective, because we have thousands of developers using our tools on a daily basis to create AR use cases, and we can see what those people are working on. The things they are doing today with AR are substantially different from what we saw two or three years ago. There are still people working on proof of concepts, but the number of people who are moving from POC to commercial grade installations – and the number of use cases we see that are no longer for two or three or five users, but 10,000 to 20,000 users – has rapidly increased in the past year.

Also, from a finance perspective, AR is no longer tapping into the budgets of the innovation units – it’s tapping into the budgets of the actual business units. That’s the ultimate sign that technologies like AR/VR are starting to take hold in the enterprise space.

Asif: There are at least three reasons why I’m very feeling positive about AR/VR. The first is the acceleration of digitalization that has taken place as a result of the COVID-19 pandemic. Many, many systems are getting digitally transformed, and digital journeys that might have taken years to complete are now on the fast track. So, the ground is really set for AR to make a move.

The second reason is that digital process management has really evolved. The journey really starts with connectivity first, then it goes to the integration, then it goes to the digital workflows. Once you have the workflow, to augment the workflow with AR is very straightforward.

The third reason is the advent and proliferation of smartphones and tablets that are loaded with the sensors and features that are required for AR/VR. These devices are now at everyone’s fingertips, ready to be used for various advanced workflows. So, really, I think the time is very, very good right now for AR/VR.

Workstation makes custom bike design easy

January 29, 2021 By Leslie Langnau Leave a Comment

While 2020 wasn’t without its challenges, we also saw innovation at its best as companies across industries incorporated emerging technologies and developed new workflows to ensure business continued to move forward amid all the chaos caused by the pandemic.

One such company is Predator Cycling, a manufacturer and designer of high-end custom carbon fiber bicycles for cycling enthusiasts and Olympic-level cyclists. Using the pandemic as a catalyst for change, Predator Cycling’s CEO and co-founder, Aram Goganian knew the latest technology gains could transform his Nashville-based business. The engineers and designers develop all of their frames and conduct all of their simulation, rendering and manufacturing processes in house. In addition, they also manufacture, build and simulate all of the machinery and equipment used to build custom bikes. They take any chance to optimize these processes throughout the design and production lifecycle.

The latest bike launch is the RF20 frame. After being in research and development stages for years, Goganian wasn’t sure the new road bike would ever see the light of day due to increasing costs of materials and the complexity of the design, which impacted bike manufacture and assembly. However, through performance and efficiency gains – including the ability to process more complex models, render, and run simulations in real time, and streamline manufacturing processes – Predator Cycling was finally able to introduce the new bike at a competitive price.

The design team uses the ThinkStation P620, equipped with the NVIDIA RTX A6000 GPU built on the new Ampere architecture. Because the team does so much custom work and each of their bikes are purpose built for each individual rider, customers need an easy way to see what their bike is going to look like and start to build an emotional connection with their bike. To do this, they use Luxion’s real-time ray tracing application Keyshot running on Lenovo and NVIDIA’s hardware to quickly load all of their existing CAD models and drag and drop all of their finishes, textures and surfaces onto the part to give the customer a realistic representation of what their bike will look like.

Seeing rendered prototypes not only provides customers with seamless options for component and finish requests that they can alter in real time, it also saves the company time and money. Prior to this, the Predator team would build physical prototypes for marketing purposes, which would take months to complete from start to finish.

They are now able to show customers renders of bikes long before they get to production or even physical prototyping. Thanks to the instant feedback they receive from customers, they can go from prototyping straight to testing. Goganian estimates that this new workflow has saved somewhere between 12-16 weeks in their go-to-market timelines.

“As I find myself deep into the design and simulation process of a component, I can be approached with an editing or simulation update task from the production line without my workflow being disrupted,” said Goganian. “I can now run concurrent simulations – from topology optimizers to thermal dynamics – all while hosting a video call with a client. As a result, I’ve saved more than 1-2 weeks in design-to-manufacturing turnaround using this workflow.”

The designers have seen performance gains of 2 to 6x across a number of the key applications they use including Luxion Keyshot, Ansys Discovery, Ansys Mechanical, Ansys CDF, and Autodesk Fusion 360. Plus, they can validate and test their designs more efficiently. For instance, the designers can implement mechanical simulations to help optimize carbon fiber orientations as well draping patterns to enhance the structural integrity and riding characteristics of their frames. They also use fluid simulation to identify any drag in order to design a frame system that will “cut through the wind.”

Moving forward, Goganian and team are exploring other ways they can take advantage of the power and flexibility of the 64-core enabled ThinkStation P620 workstation. From integrating digital twins into their design workflow and transforming their manufacturing process by using generative design and 3D printed parts to capturing and editing their own 6K video footage for marketing, Lenovo is helping Predator pave a smarter way forward.

Lenovo
www.lenovo.com/us/en

Developing a Parametric Model of a Bicycle and Human

January 25, 2021 By Leslie Langnau Leave a Comment

A general-purpose parametric SolidWorks model has been created to enable rapid evaluation of novel bicycle concepts. It can be used for ergonomics, checking clearances, aerodynamics, kinematic and degrees of freedom studies, and product visualization.

Dr. Jody Muelaner, PhD CEng MIMechE

For this study in ergonomics, a set of anthropometric models from 3D Human Model were used as the starting point. Key bicycle geometry was defined using reference geometry in individual part files, located using distance and angle mates. This geometry includes the contact points at the saddle, handlebar grips and pedals, as well as the steering axis and wheel positions. Anthropometric models were then configured to fit to these contact points, with a few additional angle and distance mates that allow further adjustment of the rider position. The included models represent 5^th percentile females, 50^th percentile males and 95^th percentile males. The model is stored as an assembly template file, enabling it to be easily reused as a layout for different design concepts. It was created for the BriefBike project, which is developing a new class of folding bicycle that will effortlessly fold into a roller-case.

Key bicycle geometry
There are three ways that references can be parametrically defined within an assembly. Using a sketch or creating reference geometry (point, axis or plane) is often more straightforward and, in the case of a sketch, allows multiple parameters to be defined in a single model tree feature. However, parameters defined in this way cannot be animated or adjusted using the Mate Controller. The Mate Controller is particularly useful within this model as it allows parameter sets for different riding positions to be stored independently of configurations. It is, therefore, possible to apply different standard riding positions to any configurations that a are created. In order to provide this greater flexibility, parameters must be defined using distance and angle mates. This means the reference geometry is first defined with a part file, and the part is then mated in the assembly file.

The parametric bicycle geometry definitions are:

Planes parallel with the Right plane, defining the width of each pedal from the centerline (Right plane).
The crank length part three axes – the bottom bracket axle and the two pedal axles. The crank length is defined by different configurations of the part file. The part is mated on the right plane with the bottom bracket axis on the Front plane and at a distance from the Top plane, representing the bottom bracket height from the ground. This leaves the pedals free to rotate.
Pedal thickness parts contain a pedal axis and a plane representing the top surface of the pedal. They are mated to the pedal axes in the crank length
A Seat part contains planes representing the seat tube angle and the top surface of the saddle. The seat tube angle is mated coincident with the bottom bracket axis and at an angle from the top plane. The top surface is mated at a distance from the bottom bracket axis.
A Bar part contains an axis to represent the handlebar. It is mated at a vertical and horizontal distance from the bottom bracket axis.
The orientation the handlebar grips is defined in terms of two sequential Euler rotations – the backwards and downwards sweep. A separate part is used to define each rotation.
Grip_Sweep parts are mated to define the position and backwards sweep, with the following constraints:
- Two translations by mating the origin coincident with the Bar axis
- The remaining translation, the width of the grips, is defined by a distance mate between the part’s origin and the assembly Right plane.
- Two rotations are constrained by mating the part’s Top plane parallel with Top in the assembly.
- The backwards sweep is the only remaining degree of freedom. It is defined with an angle mate between the part’s Front plane and Front in the assembly
Grip parts are then mated relative to the Grip Sweep part, defining the downwards sweep.
- All three translations are constrained by mating the origin coincident with the origin of Dum_Grip_Sweep
- Two rotations are constrained by setting the Front plane parallel with Front in Dum_Grip_Sweep
- The downwards sweep is the only remaining angle. It is defined with an angle mate between the Top plane and Top in Dum_Grip_Sweep
Wheels contain an axle axis and a ground plane set at the wheel radius. Different configurations are used for different wheel sizes. Certain configurations may also include basic solid geometry to visualize the tire. The wheels are mated with the ground plane on the assembly Top plane and at distances forwards and rearwards of the bottom bracket axis.
A Steering Axis part contains a plane to represent the steering axis angle and an axis to represent the line where this plane intersects with the ground. The axis is mated coincident with the Top plane of the assembly. A distance mate between this axis and the Front plane of the front wheel defines the steering trail. The steering axis angle is then set with an angle mate.

Care must be taken when using angle mates. The direction in which the angle is defined can flip when changes are made to the model, causing assembly rebuilds to fail or result in unexpected behavior. These issues can usually be avoided by defining a reference entity for each angle mate, which is not defined by default. This is normally just a one click operation, by selecting Auto Fill Reference Entity.

Kinematics of the Human Models
The human model has parts or sub-assemblies for hands, lower arms, upper arms, clavicles, head, neck, thorax, abdomen, pelvis, upper legs, lower legs and feet. These 19 rigid bodies have 114 degrees of freedom (DoF) without any joints or other constraints. Joints between the body parts are either spherical, removing the three DoF for translation, or revolute which also removes two rotations, constraining a total of five DoF.

When all of the joints are added to the body parts, the human model still has 43 DoF. Considering the kinematics of the model as a whole is, therefore, overly complicated. Luckily, it can be broken down into smaller kinematic chains that behave independently. For example, each leg forms a kinematic chain which also includes the crank, and each arm forms a kinematic chain between the shoulder joint and the handlebar grip.

An understanding of how a person should be positioned on a bike was provided by Mike Veal, who created the DIY Dynamic Bike Fitting guide.

The hip joints should align with the plane of the seat post angle.
The angle of the line between the hip and shoulder joints is typically between 45° and 55° from the horizontal. 45° to 50° is usual for a road bike and 50° to 55° is typical for a more relaxed upright position. Dutch bikes can be from 65° to 90°.
The angle of people’s feet relative to the floor is quite personal but a typical value is 15°.
The leg does not completely straighten at the bottom of the pedal stroke. Typically, the angle between the upper and lower leg does not exceeds 140°. Although some literature puts this angle closer to 150° this is due to static measurements with the foot parallel to the floor. When pedalling, the foot assumes a natural angle which reduces the extension of the leg.
Wrists allow three types of rotation and should ideally be in a neutral position for all of them:
- Flexion/Extension can be fixed at the neutral position, with the flat plane of the hand aligned with the forearm axis. It can be adjusted without significantly changing the position of the arms by rotating the hands around the axis of the grip.
- Deviation is sideways movement of the hand, towards the thumb is radial deviation and towards the little finger is ulnar deviation. The neutral position does not position a griped bar perpendicular to the axis of the forearm but rather that the third metacarpal bone is aligned with the forearm axis. One study found that a natural grip results in a mean angle of 65° between the grip axis and the third metacarpal, with the grip sweeping back as though the wrist was in 25° ulnar deviation. However, the standard deviation was 7°, due mostly to variation between individuals, suggesting significant adjustability may be desirable for this aspect of the grip position.
- Supination/Pronation: Rotation about the forearm axis is known as supination when the thumb is rotating towards the back of the hand and pronation when it is rotating towards the palm.

Kinematic chain from pelvis to head and shoulders

This section of the body is made up of the pelvis, abdomen, thorax, clavicles, neck and head. The pelvis is fixed at the saddle, but is free to rotate so that it tilts forwards. The clavicles are mated parallel with the front and top planes of the thorax, effectively forming one ridged body with the shoulders in a neutral position. Although the components could be mated in series, starting with the tilt angle of the hips and then setting the angle between each part, this would make it hard to set an overall lean angle. A reference part is therefore introduced with a plane that defines the lean angle. This part is mated with an axis through the hip joints and with an angle mate relative to the assembly Top plane.

A symmetry mate is used to apportion half of the forward lean to the pelvis tilt. The Front plane of the pelvis is the plane of symmetry. The seatpost angle and the forward lean planes are symmetric about it. The abdomen is mated parallel with the dummy lean plane, and the shoulder joints are set to be coincident with the forward lean plane.

Kinematic chain from hip joint to bottom bracket

Each leg can be considered separately, as a kinematic chain consisting of the crank, pedal/foot, lower leg, and upper leg.

These four bodies have 24 DoF, which are reduced by revolute joints at the knee, pedal axis and crank axle, and spherical joints at the hips and ankles, to leave three DoF:

Crank position is intentionally unconstrained to allow pedaling motion.
Foot angle from the floor varies between individuals but the toes are typically pointed downwards by an angle of approximately 15 degrees.
The spherical joints at the hip and ankle also allow the leg to rotate about an axis between these joints, so that the knee moves in a circular motion. Assuming the leg does not lock out into a fully extended position, this can be constrained by making a point on the knee joint coincident with the plane defining the pedal width.

Kinematic chain from shoulder to handlebar grip

The kinematic chain for each arm consists of the upper arm, lower arm and hand. These three bodies have 18 DoF, reduced to just two DoF by spherical joints at the shoulder and wrist, and revolute joints at the elbow and the grip of the hand around the bar. The remaining two DoF can be considered as:

Rotation of the hand around the handlebar
Rotation of the whole arm so that the elbow moves in a circular path about the axis between the shoulder and wrist joints

Setting the wrist to neutral flexion removes one DoF. This can be achieved by mating the hand’s top plane parallel with the forearm axis. There are several possible ways to remove the remaining DoF. It was found that the most practical and stable is with a distance mate between a point on the elbow joint and the Right plane of the assembly.

Adjusting and configuring the model

The assembly template contains three different anthropometric models, all configured as described above. These models represent a 5th percentile female, 50th percentile male and 95th percentile male. They can be activated by simply suppressing or suppressing the associated folders in the feature tree.

Different pre-configured body positions are also included. These are defined using a Mate Controller for each percentile model.

Conclusions

The bicycle and human model template enables the rapid evaluation of novel design concepts. Applications include ergonomic studies, mechanical clearance checks, aerodynamic simulation and product visualization. The simple parametric definitions allow easy adjustment of all the relevant variables in the bicycle geometry and rider position.

Introducing hyperMILL 2021.1 CAD/CAM software suite

January 19, 2021 By Leslie Langnau Leave a Comment

OPEN MIND Technologies AG, a leading developer of CAD/CAM software solutions, has introduced its latest hyperMILL 2021.1 CAD/CAM software suite which offers users new and enhanced features for efficient 3D, 5-axis and mill/turn machining. Key innovations include a new “Interactive Edit Toolpath” capability which enables toolpath editing after initial toolpath generation. Especially productive for optimizing tool and mold making, the new toolpath editing feature is easy-to-use, and offers programmers the flexibility to adapt toolpaths by trimming and removing sequences accordingly for component conditions.

To streamline access to Product Manufacturing Information (PMI) and metadata, hyperMILL 2021.1 offers a new import function that retrieves face quality information and metadata when importing CAD data from neutral or native formats, and attaches data to the imported faces in hyperCAD-S, making the information available to hyperMILL and its machining processes.

A new 5-axis Radial Machining strategy allows bottle shapes and similar cavities to be easily and efficiently programmed in hyperMILL, resulting in high-quality surface finishes. Using a new radial projection method, toolpaths are calculated quickly so that the most productive machining strategies can be applied. For optimal high precision machining, hyperMILL “High Precision Surface Mode” and “Smooth Overlap” strategies can also be applied when using 5-axis Radial Machining, ensuring the best surface quality and clean transitions.

For high-quality surface finishes and to simplify programming when flank and point milling blades, hyperMILL 2021.1 offers several enhanced Multi-Blade strategies. No longer do blade surfaces need to be ruled for accurate programming results. Using the enhanced strategies, any number of surfaces are permitted for the suction and pressure sides, making it exceptionally easy to extend blade surfaces. Fillets with a variable radius are supported in the latest version of hyperMILL. Also, for high accuracy, the enhanced Multi-Blade flank milling strategies result in smaller deviations on the suction and pressure side, and offer improved tool guidance along the upper boundary in the edge area.

hyperMILL 2021.1 provides a new SIMULATION Center for generic NC data in turning and milling operations. Modeled after the hyperMILL VIRTUAL Machining Center, the new, modern SIMULATION Center is integrated in hyperMILL, and offers an intuitive operating environment for faster simulation, independent collision checking and extensive analysis functions.

Additional new features in hyperMILL 2021.1 include a new, easy-to-use “XY Optimization” command for optimal 3D profile finishing. In the hyperMILL Mill-Turn Module, the high-performance mode has been integrated into a 3-axis simultaneous roughing strategy, combining the advantages of HPC and simultaneous turning. For the simplified alignment of components, hyperCAD-S has a new “Align Best Fit” machining command that allows like-geometry components to be aligned with one another using defined pairs of points.

OPEN MIND Technologies AG
www.openmind-tech.com/en

The latest release of Simcenter 3D 2021

January 14, 2021 By Leslie Langnau Leave a Comment

Siemens Digital Industries Software announces the availability of the latest release of Simcenter 3D software, part of the Simcenter portfolio of simulation and test solutions. Simcenter 3D and the Simcenter portfolio are part of the Xcelerator portfolio, Siemens’ integrated portfolio of software, services and application development platform. In the 2021 release, Simcenter 3D continues to further improve its powerful unified and shared engineering platform for all simulation disciplines to help users gain full value of the benefits that simulation provides in terms of cost, speed and impact to innovation. Introducing new enhancements to the AI (Artificial Intelligence) driven user experience, new simulation types as well as refinements in accuracy and enhanced performance speed, Simcenter 3D 2021 can help companies understand true performance of their designs early in their development process.

Evaluate sound performance with auralization: New in Simcenter 3D, pressure results as a function of time (bottom right) can be played as an audio file.

In many applications, product innovation includes the engineering of the advanced material used in them, which is why new materials are being introduced into the market at unprecedented speed. Cracking is a very important consideration for advanced materials, however micro and meso cracking in advanced materials is difficult to model with the finite element method. Simcenter 3D now includes full representative volume element (RVE) separation and 2D and 3D automatic insertion of cracks or cohesive zones in materials. Macro and microstructural models now allow for full mesh separation for a crack to propagate completely through a material.

“Simcenter Multimech allows us to model microstructural cracks and determine how they would affect the overall part,” states Neraj Jain, group leader in simulation and engineering at the DLR Department of Ceramic Composites and Structures. “Using this tool, we can actually see where a crack is developing, how the crack will change our material, and how it will affect the final microstructure of the material.”

New to Simcenter 3D is an auralization post-processing tool that allows users to listen to simulated pressure results to evaluate sound quality. This allows acoustics engineers to actually hear the noise produced from various vibrating components or products as opposed to having to visually evaluate through charts or graphs.

Simulation-driven design can drastically lower the time it takes to bring a product to market. For this reason, Simcenter 3D’s thermal analysis capabilities have been scaled into a vertical solution for mold designers and design engineers. The new NX Mold Cooling product uses Simcenter 3D technology to allow designers to rapidly set up and simulate the thermal performance of an injection mold insert directly in NX as they are designing the mold. This allows for easy and rapid thermal analysis of injection mold designs without having to wait for expert analyst feedback.

How to simulate multiple physical forces

December 15, 2020 By Leslie Langnau Leave a Comment

Multiphysics simulations are becoming more complex, and take place earlier in the design cycle

Jean Thilmany, Senior Editor

Using multiphysics simulation software in tandem with design keeps engineers from following hypotheses that lead to dead ends. Of course, that significantly cuts development time, especially for complex products on which many physical forces come together: for instance, solid-state cooking ovens and the sonar on unmanned, undersea drones.

If the examples seem oddly specific, it’s because engineers who develop both types of applications described how their teams use Comsol multiphysics software at the Comsol Conference held online in October. The analysis software simulates how multiple, real-time physical forces would affect design.

Because Lauren Lagua works for a sonar team primarily funded by research and development dollars, “We don’t have a lot of time to spend on design and implementation,” she says. Lagua is an acoustic engineer with The Northrop Grumman Undersea Systems group. As a mechanical designer and acoustic analyst, she integrates tests equipment for undersea sonar systems and payloads for unmanned underwater vehicles (UUVs).

To make the most of funds, her team created a workflow for quick test-and-verify design. For transducers, their design method calls upon swift, but multiple rounds of prototyping and verifying. All the while, they must ensure the transducer can be rapidly manufactured. The transducers convert variants in pressure, brightness, or other physical qualities into an electrical signal.

Though Lagua’s team solves for electrical and acoustics tests early in the initial design stage, that doesn’t mean they’ve perfected design right off the bat. The process involves many design changes and modifications—or as Lagua puts it, “around and around we go.”

Of course, the models begin with design and CAD, but using Comsol for analysis in conjunction with design is the secret sauce that speeds the process, Lagua says. Because they simulate as they design, they get to the end result much faster than if they had to hand off the model to a dedicated analyst and await results.

“It allows us to quickly iterate on a design, to try out designs, to identify issues early on in the design phase and mitigate and fix them before they become bigger issues,” Lagua says.

“Say I’m building a transducer made of some type of material and I bond it to an electric substrate and see a bubble,” she says. “I can make a hypothesis about what failed. I then change the design, and test it with Comsol and compare it to our test events. We find issues with models and correct them quickly.”

“We’re able to design, prototype, test and verify a design sometimes in as little as a week,” she adds.

New materials and tank testing
Because it simulates how the material from which the object is made affects performance, the analysis software also helps engineers establish the material properties the transducer needs. They also use it to test prospective new materials, Lagua adds.

“We figure out if we have to change the design based on the prototype and maybe changing the materials is part of that,” she adds.

“The challenge is: materials vendors don’t always give you all their materials properties. And you need them,” Lagua says. “So, we use Comsol to model the materials.”

The researchers do this by plugging the materials properties they do have into the simulation and analysis software. They also find the materials they deem most similar to the one they are using and put those material properties in those as well.

“So, if I’m experimenting with a new polyurethane, it comes down to what we know about polyurethane in general. Then we look at the differences in data between the model and the test information and change and tweak the information we do have,” she says.

What is the test information she refers to?

Well, of course, the team needs real-world measurements to verify against and plug into the analysis software. These tests are also how they verify material properties against the information they have. They create model prototypes and study them, to study how that prototype would perform in the field.

Acoustical test events, as they’re called, take place at a specialized underseas testing facility that simulate open-water testing. The facility her team uses features a 50-foot-diameter, 300,0000-gallon tank lined with redwood for sound baffling. Onsite test and measurement equipment are placed on the prototype to gather immediate feedback and readings.

“It’s the closest an engineer gets to measuring how a device will act deep in the ocean—without actually taking the device deep in the ocean,” Lagua says.

“We take the results we get from the facility and bring them back into Comsol,” she says. “Then, with Comsol, you can run the same acoustic testing you do at an open-water tank,” she adds. “We have that information in our analysis software.

“Then we can and relate it back to our earlier test data and verify and tweak our model,” she adds. “We figure out if we have to change design or maybe change materials to optimize transducer performance.”

The engineers optimize transducer performance-based how that particular transducer’s end use.

“For example, we’ll optimize for acoustical performance: the highest level of sensitivity to capture the lowest amount of sound while having a large, broadband frequency coverage to hear over a very large range of frequencies,” Lagua says. “We’re really optimizing for multiple things in parallel.”

At the conference, Lagua shared one of the systems created through her group’s design iteration process: a µSAS (pronounced “micro-sas”) sonar to be affixed on UUVs. The micro-sas is small in size, weight, and power.

The U.S. Department of Defense is adapting Unmanned Underwater Vehicles, which function, in some ways, as a submarine without personnel aboard.

The interferometric synthetic aperture sonar produces high-quality, 3-D images, enables longer sorties, and higher area coverage rates for UUV missions, says Alan Lytle, vice president of undersea systems at Northrop Grumman. The sonar takes 2-D sas and 3-D bathymetric images.

The UUV vehicles carry six sonar systems, which can be preprogrammed for the needs of the mission, Lagua says.

Northrop Grumman Corp.’s announced in February itµSAS (pronounced “micro-sas”) will be integrated onto L3Harris Technologies’ Iver4 Unmanned Undersea Vehicle, pictured for a 12-month test period, as part of the Defense Innovation Unit’s Generation Small-Class UUV program.

“We have sonars on both sides of the vehicle so you can look at images from two different aspect ratios and interpret them in 3-D just as your eyes would,” she adds.

“Because of its very small scale we were constrained in size, weight and power for our system,” she says. “We wanted the best acoustics possible at the smallest scale while conserving as much energy as much as possible.”

The group tested their late prototypes by sending them down to take imagery of a ship that sank on the bay near their Maryland facility.

Engineers at Northrup Grumman used acoustics simulation technology to develop the µSAS sonar mounted on the L3Harris Technologies’ UUV.

“In the 3-D image you can see the ship is in a big trough and the front is sticking out of a hole,” Lagua points out. “You wouldn’t be able to see that in 2-D.”

While she did not say exactly what sonar technology her team worked on, in February, Northrup Grumman announced that the company’s µSAS interferometric synthetic aperture sonar will be integrated onto L3Harris Technologies’ Iver4 UUVs.

The Iver4 UUV weighs 200-pounds, is nine inches in diameter and 99-inches long. Integration of synthetic aperture sonar on UUV of this diameter represents a significant step forward for the operational capability of small-class vehicles, according to Lytle.

The next-gen microwave oven
Then we veer to quite a different use for simulation in the early (and late) design stages. Illinois Tool Works (ITW) Food Equipment Group uses Comsol to analyze and simulate new heating methods for their solid-state ovens. The ovens, intended for commercial use, can cook a variety of foods, all at the same time and their different temperatures, says Chris Hopper, radio frequency systems engineer at ITW. He also spoke at the Comsol conference.

Solid stage ovens differ from convection microwaves, which use the same magnetron technology developed for radar in World War II. A “regular” microwave oven uses the open-loop magnetron system to heat foods.

Microwave ovens use radio frequency technologies developed in World War II. ITW Food Technologies is working on solid state ovens that would be an upgrade to the microwave. (Photo source: ITW Food Equipment Group)

But magnetron-based systems have many limitations, including low power and phase control, short lifetime, and high-voltage power supplies, Hopper says.

On the other hand, RF solid-state cooking features a closed-loop feedback system that can adapt to various loads and measure the food’s properties at any time during the cooking process, Hopper says.

“With solid-state, you can vary the power, measure what goes into the cavity and is coming out, and you can teach the oven how to intelligently respond over time to the feedback it gets,” he says.

“A magnetron can last from 12 to 18 months, but with solid-state power, the lifetime can be amplified for many years,” Hopper adds. “And the performance doesn’t degrade over time.”

“But before you start building, you want to investigate basic physical phenomena because we’re talking about a metal box with multiple phenomena,” he says.

Hopper’s team uses Comsol to iterate on the design using LiveLink for MatLab. The software allows them to synchronize their model with Comsol Multiphyics and define geometry, run multiphysics simulations, and optimize the model accordingly. For instance, they use simulation to study their model’s heating patterns.

“We look at what the presence of the food changes in terms of interference, hot and cold spots, and other qualities,” he says. “How does changing the phase affect the food itself?

“We don’t need it to be exact,” Hopper adds. “We’ll look at the accuracy once we’ve established that the simulation represents real-life experiments accurately,” he says.

“The culmination of our work is to develop algorithms,” Hopper says. “We can study hundreds of combinations of different phases in different sources. We can look at the data behind this and test and train models mainly worthy of further development for an algorithm.

“With simulations, we can phase out what wouldn’t work for outcomes we’re interested in,” Hopper says. “It saves us a lot of time because we don’t go down dead ends for algorithm development that may not work out.”

To study each phase and frequency combination separately would take weeks using testing equipment in a lab, he adds.

The virtual simulations also cuts labor costs in an unusual way, as the engineers run fewer experiments with actual oven prototypes and the food they cook.

To make those simulations available to product specialists, who don’t need or want to see the fine print, the ITW engineers created an application from the Comsol data. Product specialists download it, view it, and offer feedback based on their own experience, Hopper says.

“We have a chef here who is responsible for bringing value to our customers,” Hopper says. “There are certain questions he wants answered and, through the app, he doesn’t want to try it out in the kitchen many times. “He can look at temperature, airspeed, time and determine the parameters of the food he can cook and how the food will change as the result of those things.

In the end, many companies are learning that multiphysics simulation can be as important as model creation itself, says Bjorn Sjodin, Comsol vice president of product marketing. He calls the trend “the democratization of simulation.”

“More engineers are using simulation,” he says. And he expects the trend to continue.

The capability to simplify complex analysis and simulation by making application expands the reach beyond engineers.

Soon, a sales representative may be demonstrating oven phase and frequency to a potential customer. After all, “there’s an app for that.”

COMSOL
www.comsol.com

Latest release of IronCAD helps move 3D designs to production faster

November 30, 2020 By Leslie Langnau Leave a Comment

IronCAD, a 3D CAD productivity platform for metal fabricators and custom machinery manufacturers, announces the release of IRONCAD 2021, which contains many exciting improvements and capabilities that help customers drive innovation and move to production faster.

Every year, IronCAD global customers submit feedback with enhancement improvements that matter the most for increasing the design productivity or improving the 3D to the production drawing process. In addition to user feedback, after having endured an unprecedented year, it was essential to keep IronCAD users working and ensure that productivity is the way of life for their business with the release of IRONCAD 2021. In this year’s release, the main focus was on improving productivity to help create designs and production drawings faster. With this in mind, the goal for IronCAD 2021 was to improve the performance from 3D to the 2D detailing stage, improve the detailing user experience, continuous quality improvement, as well as improvements that make the design process more productive.

Examples of the new improvements in IRONCAD 2021 are:

Improved IntelliShape Handles for Increasing Performance – New behaviors to quickly symmetrically size shapes with a right-click drag, quick access to handle values attached to the cursor, and more powerful snap options to get precise locations from a point or center point that allow users to design faster with IronCAD shapes.

User Interface Improvements to Enhance the User Experience – Easier access to multiple catalogs, direct feedback on face and edge length/area information, direct selection access to assemblies, parts, features, and faces in the current selection viewing direction, TriBall® shortcuts to reduce steps in repeating copy/link commands, and many more to improve the user workflow in design.

Sheet Metal Design Improvements and Accessibility – Improved processes in selecting stocks, automatic bend alignment on angled sheet metal stock for creation and updates, and accessing common commands for sheet metal bends to speed up the design and editing of sheet metal.

Overall 2D Technical Drawing User Improvements – 2D Annotation catalog for common annotations, 2D Template catalog to quickly change templates for drawings, new tools for revision clouds and ISO Tolerance Codes for dimensions, improved bulk view creation options to automatically generate drawing layouts that speed up the 2D detailing, and many other improvements to enhance the detailing process for production designs.

Communication Improvements for Sharing – The free Share 3D viewer has an improved user interface and support for measure and textures while supporting large geometry data sets that allow users to communicate 3D designs faster and easily with customers on any tablet or laptop device on any platform with an HTML5 supported browser.

“In a year with major uncertainty and with businesses and people across the globe impacted in unique ways, IronCAD worked closely with our customers to shape the 2021 release to meet their current needs and to help plan for future capabilities for remote working and collaboration,” stated Cary O’Connor, V.P. of Marketing for IronCAD. “IronCAD remains committed to focus on our customer satisfaction and to deliver solution to assist our customers in remaining competitive and innovative in these challenging times. IronCAD 2021 is our first stepping stone in this direction and more solutions are being developed with our teams to deliver innovative solutions to our customers.” he continued.

Try the newest, updated version of IronCAD now for free by navigating to https://www.ironcad.com/free-online-trial/ to start your free online trial.

IronCAD
www.ironcad.com

Experience a seamless workflow from CT scan to full statistical analysis

November 18, 2020 By Leslie Langnau Leave a Comment

Volume Graphics is extending data export from its non-destructive testing and analysis software based on industrial computed tomography (CT) to the statistics software Q-DAS qs-STAT. This close cooperation between Volume Graphics and Q-DAS, which are now united under the roof of Hexagon, takes automation to the next level: statistical evaluations can now be fully integrated into a CT scan data analysis workflow.

Many users want to statistically analyze the quality data of their scanned components. This is often a basic requirement for automated applications. Q-DAS solutions, such as the Q-DAS qs-STAT package, are considered to be the de-facto standard for statistical analysis. Volume Graphics had already implemented an option for data export to the Q-DAS software in the earlier Release 3.3 of VGSTUDIO MAX. But now, as an integral part of Hexagon Manufacturing Intelligence, Volume Graphics and Q-DAS are working on making the data exchange between computed tomography and statistics even tighter.

Volume Graphics software transfers not only the measured values, but also an image of the CT-scanned component (in this case a cast part) into the Q-DAS qs-STAT statistics software package. This enables the dimensions to be clearly linked to the corresponding detail in the report. Visual support makes it easy for users to identify critical processes quickly, as well as weak points and deviations; green bars show time-series of measured values for specific features of the object. Summary graphics like those shown here are an essential tool for the definition of corrective actions needed to optimize a manufacturing process.

The first result of this collaboration is the option to also export 3D representations of components or measured features to Q-DAS software. This option was recently introduced with version 3.4.3 of VGSTUDIO MAX, VGMETROLOGY and VGinLINE. In practice, the user simply marks the relevant box in the export mask. The software then adds the corresponding part image to the data to be exported. Users can make their reports, which they define and access with Q-DAS qs-STAT, even more transparent. They can immediately see which measurement series belongs to which detail (see image).

“Currently, it is possible to export 3D component representations from our coordinate measuring module using measurement data,” says Johannes Knopp, Product Manager Automation & Inline at Volume Graphics. “But progress does not stop there. Our development department is already working on including additional results from the array of gray-value-based material analyses, such as defect analyses, in the export functions. The aim is to realize a seamless, fully automated workflow for all CT quality data.”

CoreTechnologie presents a new 3D printing software version at Formnext Connect

November 3, 2020 By Leslie Langnau Leave a Comment

The German-French software manufacturer CT CoreTechnologie will present a comprehensively revised version of the 4D_Additive software at the virtual Formnext Connect tradeshow from 10th to 12th of November 2020.

Since the trade fair this year is taking place online only, interested parties can get a demonstration of the software on the virtual booth of CT and ask questions directly to the software specialists. By this means, the concentrated know-how of CoreTechnologie for solving the tasks and challenges of additive manufacturing is available to the visitor.

On 10th and 11th November 2020 each day at 10am and 2pm, Expert Sessions will be held, in which an overview of the software will be presented and the extensive innovations of the software will be explained in detail. New tools for the creation of 3D surface textures as well as structures for lightweight components and for saving the models in high-precision step-format will be explained and shown live. The Expert Sessions will also be available as video in the Formnext media library.

CoreTechnologie
www.coretechnologie.com/products/4d-additive