VERICUT Users' Forum

Susan

Member

From: Singapore

Registered: 2006-09-18

Posts: 2

Problem when using AUTODIFF

Hi,

I met a problem when using AUTODIFF. I tried to measure the gouge and excess value after cutting a NURBS surface. Different representing models for the stock and the design were utilized as:

1st case: both the stock and the design are .igs file and are imported to the Vericut with tolerance 0.005 mm.
2nd case: the stock is .igs file, while the design is .stl file which is produced in UG with both Triangle Tol and Adjacency Tol are 0.005 mm.

After cutting with the same tool-path, AUTODIFF(Comparison Method: Surface) were utilized  to find the gouge and excess. The results shows the totally different gouge/excess distribution in 2 cases, i.e. no any overlapping in 2 cases. The gouge/excess location in 1st case is shown no gouge in 2nd case.

My questions are:

1. Why does this happen?
2. What is the best way to represent the models for cutting a NURBS surface?
3. What is the difference between the surface comparison method and the sold comparison method (in AUTODIFF)?

Thanks in advance for you reply.

SergeV

Senior Member

From: Irvine, CA

Registered: 2004-10-08

Posts: 507

Website

Re: Problem when using AUTODIFF

Hello Susan,

many factors will affect the quality of AUTODIFF, but the most important is the Cutting Tolerance. The best results are achieved with the Cutting Tolerance based on the Stock instead of the Tools. The cutting tolerance value should be at least equal to the gouge value you want to detect.

On small parts it might not be a problem but on large parts, it might not be possible to achieve such a small cutting tolerance. The solution is to do AUTODIFF (comparaison method = Solide) by region. This will allow you to do more precise analysis on a small area of the part instead of the full part. By checking the button "Improve Cut Stock Tolerance" the selected are will be recalculated with a cutting tolerance small enough to detect the specified gouge/excess tolerance.

In 6.1, the constant gouge check can also be used during the simulation to detect very small gouges. Instead of comparing the cut stock with the design, it analyses the cutter against the design, this requires less CPU and is very precise. It also does not affect the simulaiton speed.

Finally, it is possible that a difference between the models imported in VERICUT from an IGES (VERICUT Polygone format) could differ from the STL exported from UG. Every system has its own algorithm to create faceted models. Specially with such a small tolerance.