What is Patch Planning?
ICEMSurf class-A surface models are built from a collection of individual quadrilateral (4-sided) Bezier surfaces, known as patches. In areas where non-quadrilateral surfaces are necessary then the patches will be trimmed. These are known as faces. A collection of patches and faces is known as a quilt (as in patchwork quilt).
Patch planning is the art of dividing a 3d sculptured shape into a quilt of patches and faces.
Patch planning is the art of dividing a 3d sculptured shape into a quilt of patches and faces.
Objectives of Patch Planning
What are we trying to achieve with our patch plan?
1. Accuracy
We need to create an class-A surface model that is accurate to the scan data and to the designer's intent.
2. Simplicity
The 3D surface model should be efficient in terms of the surface definition required for a given shape. This means that we must try to minimise the number of patches/faces and we should use the lowest patch orders (i.e. the number of control points) possible. Such lightweight models will give smoother surface results as well as making modifications easier and viewing faster. Lightweight models become very important if they are to be part of a large assembly such as a car interior.
3. Surface Quality
We want our ICEMSurf surface model to be of the highest quality. Highlights should be as clean and smooth as possible and our patches and faces should be matched to the highest level (from G0 to G3 as required).
4. Uniform Control Points
Ideally our surface control points should be uniformly positioned. This helps when matching the patches together. Some degree of control point bunching is acceptable, however, as, if done properly, this can reduce the number of control points that we need.
5. Downstream Operations
Once released, the ICEMSurf model will usually be exported into a downstream CAD system, such as Catia V5, where CAD engineers will carry out further operations, such as offseting, on the surface model in order to develop the fully engineered component as a Solid CAD model. So our ICEMSurf model should be structured in a way that makes these downstream CAD processes as easy as possible.
Patch Plan Development
If you're new to class-A surfacing then I would suggest that you develop your patch plan by sketching over plots of the scan data or copies of the designer's sketches (don't draw on the designer's originals - he probably won't be too pleased!). If you speak nicely to the clay modellers they may well let you put tape on the clay to help define your patch plan in 3D. It's always useful to find out how the clay modellers developed the clay in any case as clay and surface modelling are similar activities.
As you gain experience you'll find that you'll be able to develop the patch plan by throwing curves and surfaces into ICEMSurf directly. Even then, there are always difficult areas (especially junctions of fillets) where a sketch beforehand can help.
The stages to developing a successful patch plan are:
1. Identify Edges, Holes and Breaks
These are obvious features of the geometry. In some cases our patch/face edges will extend beyond or over these features, but in other cases these define where our patch/face edges should be. We'll discuss this later.
2. Identify Fillets
Small fillets will be obvious. In the first stages of the patch planning we can pretend that the fillets are not there i.e. if we extend our block surfaces they will meet at a hard intersect - the "theoretical intersect". There are times when we will want to crash the block surfaces through each other; in other cases we will want our surface edges to coincide at the theoretical intersect.
3. Identify Block Surface Regions
The regions where the block surfaces lie are the areas left when we have picked out the features described above.
4. Define Patch Structure
The final stage is to split the block surfaces into individual patches and faces. There are two aspects to this:
(i) the orientation of the patches/faces and
(ii) where to split the patches/faces.
What are we trying to achieve with our patch plan?
1. Accuracy
We need to create an class-A surface model that is accurate to the scan data and to the designer's intent.
2. Simplicity
The 3D surface model should be efficient in terms of the surface definition required for a given shape. This means that we must try to minimise the number of patches/faces and we should use the lowest patch orders (i.e. the number of control points) possible. Such lightweight models will give smoother surface results as well as making modifications easier and viewing faster. Lightweight models become very important if they are to be part of a large assembly such as a car interior.
3. Surface Quality
We want our ICEMSurf surface model to be of the highest quality. Highlights should be as clean and smooth as possible and our patches and faces should be matched to the highest level (from G0 to G3 as required).
4. Uniform Control Points
Ideally our surface control points should be uniformly positioned. This helps when matching the patches together. Some degree of control point bunching is acceptable, however, as, if done properly, this can reduce the number of control points that we need.
5. Downstream Operations
Once released, the ICEMSurf model will usually be exported into a downstream CAD system, such as Catia V5, where CAD engineers will carry out further operations, such as offseting, on the surface model in order to develop the fully engineered component as a Solid CAD model. So our ICEMSurf model should be structured in a way that makes these downstream CAD processes as easy as possible.
Patch Plan Development
If you're new to class-A surfacing then I would suggest that you develop your patch plan by sketching over plots of the scan data or copies of the designer's sketches (don't draw on the designer's originals - he probably won't be too pleased!). If you speak nicely to the clay modellers they may well let you put tape on the clay to help define your patch plan in 3D. It's always useful to find out how the clay modellers developed the clay in any case as clay and surface modelling are similar activities.
As you gain experience you'll find that you'll be able to develop the patch plan by throwing curves and surfaces into ICEMSurf directly. Even then, there are always difficult areas (especially junctions of fillets) where a sketch beforehand can help.
The stages to developing a successful patch plan are:
1. Identify Edges, Holes and Breaks
These are obvious features of the geometry. In some cases our patch/face edges will extend beyond or over these features, but in other cases these define where our patch/face edges should be. We'll discuss this later.
2. Identify Fillets
Small fillets will be obvious. In the first stages of the patch planning we can pretend that the fillets are not there i.e. if we extend our block surfaces they will meet at a hard intersect - the "theoretical intersect". There are times when we will want to crash the block surfaces through each other; in other cases we will want our surface edges to coincide at the theoretical intersect.
3. Identify Block Surface Regions
The regions where the block surfaces lie are the areas left when we have picked out the features described above.
4. Define Patch Structure
The final stage is to split the block surfaces into individual patches and faces. There are two aspects to this:
(i) the orientation of the patches/faces and
(ii) where to split the patches/faces.