AnsweredAssumed Answered

What does 100% really mean for "Untrim Surface"?

Question asked by Dan Pihlaja on Jan 8, 2019
Latest reply on Jan 14, 2019 by Thomas Voetmann

Like • Show 4 Likes4
Comment • 9

OK, here is the scenario:

I have this little snippet of a surface on a solid body that looks like it was created by a chamfer. This is an imported part and I just wanted to extend the surface so that it crosses my plane that I am working on.

So, I offset the surface 0" to effectively copy the surface.

Then I select "untrim surface" to 100% and select it.

However, this does not extend the surface as far as I want it to.

So, I thought about it....and I wondered if I could untrim it again.....

I would think that I couldn't...because I previously untrimmed it to 100%.....That is maximum...right?

Sure enough....selecting the surface that I JUST untrimmed and untrimming it again.

Now, it looks like a half truncated cone.

What if I keep selecting it?

Untrim # 4:

And, apparently, I can keep going, potentially extending the height of the cone infinitely.....

So....what does "100%" really mean? How does the program know when to stop?

I looked at the surface areas of the different surfaces and there really isn't a pattern.

Some of them are almost exactly 4 times the surface area of the previous one...some are almost exactly 6 times.

Matt Lombard Jan 8, 2019 12:30 PM

Correct Answer

For some surfaces, it has a meaning. Like a sphere. 100% is easy to visualize for a sphere. A cone is bounded on one end, but on the other it could be infinite, as with other analytical types. But it doesn't really work that way.

You can use Extend where it works better, and sticks to the topological type.

100% has more meaning for a bounded general NURBS surface, where the math only goes so far. Extend can extrapolate, but the curvature gets unpredictable..

So from these experiments, for analytical types, 100% Untrim could/should be predictable, but isn't. For general spline surfaces, 100% untrim appears to be fully predictable.

See the reply in context

No one else had this question

Outcomes

Paul Salvador Jan 8, 2019 11:54 AM

Hello Dan,.. very good question!

I always wondered as well,.. it would seem "relative" to the face/boundary/area?

It would be GREAT if the surface topology was recognized within the feature as either analytical or as nurbs surface and it gave us a choice,.. and, your cone example would have a single option for full 100% untrim within a analytical boundary limit.

Actions

Dan Pihlaja @ Paul Salvador on Jan 8, 2019 12:35 PM

Paul Salvador wrote:

Hello Dan,.. very good question!

I always wondered as well,.. it would seem "relative" to the face/boundary/area?

It would be GREAT if the surface topology was recognized within the feature as either analytical or as nurbs surface and it gave us a choice,.. and, your cone example would have a single option for full 100% untrim within a analytical boundary limit.

So, how would this analytical boundary limit be determined? (I don't understand nurbs surfacing enough to be able to comment on that part of it ).

Would this be something that we would define in the context of the untrim command? Or would the program automatically put a limit on it?

At first I thought that the "100%" meant that the surface area would be 100% larger....but that is not the case.

In the CATIA Forums, it seems to imply that their program recognizes when untrimming it would result in an infinite surface and then limits it by the bounding box.

Actions

Paul Salvador @ Dan Pihlaja on Jan 8, 2019 12:54 PM

Dan,.. SW would have to have a limit with some of the geometry, the cone wraps to 360 but the base could extend for infinity.. and I would guess it depends on the analytical geometry (plane, cone, sphere, ellipsoid, hyperboloid, paraboloid..) to set the limit maybe first set within the work space/area or some ratio of existing area... and the external limit would be 1km sq?

Actions

Roland Schwarz Jan 8, 2019 11:55 AM

That is a great question.

I was introduced to the concept of untrim in UG(NX). NX remembers the original full untrimmed definition and untrims to that.

SW is kinda arbitrary. 0 equals 100%, unless it means minimal untrim. 100% = 100% unless it means 200%. There's an elegant randomness to the whole thing.

Actions

Dan Pihlaja @ Roland Schwarz on Jan 8, 2019 12:09 PM

Roland Schwarz wrote:

That is a great question.

I was introduced to the concept of untrim in UG(NX). NX remembers the original full untrimmed definition and untrims to that.

SW is kinda arbitrary. 0 equals 100%, unless it means minimal untrim. 100% = 100% unless it means 200%. There's an elegant randomness to the whole thing.

In this case, there would be no "full untrimmed definition" because this is an imported model. I would assume that this data would not have come through the translation.

So the question still stands.....why did the first untrim stop where it did? Why not create the cone in the 3rd untrim function?....and why does it go to that particular boundary?

Actions

Frederick Law @ Dan Pihlaja on Jan 8, 2019 12:14 PM

First one stop at 90 deg?

Second 180 deg?

Actions

Dan Pihlaja @ Frederick Law on Jan 8, 2019 12:30 PM

While one side did indeed stop at the next main plane, the other side did not:

Although, this 2nd untrim isn't 90 degrees from the 1st one:

Actions

Matt Lombard Jan 8, 2019 12:30 PM

You can use Extend where it works better, and sticks to the topological type.

100% has more meaning for a bounded general NURBS surface, where the math only goes so far. Extend can extrapolate, but the curvature gets unpredictable..

So from these experiments, for analytical types, 100% Untrim could/should be predictable, but isn't. For general spline surfaces, 100% untrim appears to be fully predictable.

Attachments

analytical.SLDPRT.zip231.5 KB
spline.SLDPRT.zip326.1 KB

Actions

Thomas Voetmann @ Matt Lombard on Jan 14, 2019 2:58 AM

It is interesting, that if you use untrim/100% one more time on the spline surface, it will extend the surface linear. I can not however figure out how much.

Actions

9 Replies

Paul Salvador Jan 8, 2019 11:54 AM
Mark Correct
Correct Answer
Hello Dan,.. very good question!

I always wondered as well,.. it would seem "relative" to the face/boundary/area?

It would be GREAT if the surface topology was recognized within the feature as either analytical or as nurbs surface and it gave us a choice,.. and, your cone example would have a single option for full 100% untrim within a analytical boundary limit.
Like • Show 5 Likes5
Actions
- Dan Pihlaja @ Paul Salvador on Jan 8, 2019 12:35 PM
  Mark Correct
  Correct Answer
  Paul Salvador wrote:
  
  Hello Dan,.. very good question!
  
  I always wondered as well,.. it would seem "relative" to the face/boundary/area?
  
  It would be GREAT if the surface topology was recognized within the feature as either analytical or as nurbs surface and it gave us a choice,.. and, your cone example would have a single option for full 100% untrim within a analytical boundary limit.
  So, how would this analytical boundary limit be determined? (I don't understand nurbs surfacing enough to be able to comment on that part of it ).
  
  Would this be something that we would define in the context of the untrim command? Or would the program automatically put a limit on it?
  
  At first I thought that the "100%" meant that the surface area would be 100% larger....but that is not the case.
  
  In the CATIA Forums, it seems to imply that their program recognizes when untrimming it would result in an infinite surface and then limits it by the bounding box.
  Like • Show 2 Likes2
  Actions
  - Paul Salvador @ Dan Pihlaja on Jan 8, 2019 12:54 PM
    Mark Correct
    Correct Answer
    Dan,.. SW would have to have a limit with some of the geometry, the cone wraps to 360 but the base could extend for infinity.. and I would guess it depends on the analytical geometry (plane, cone, sphere, ellipsoid, hyperboloid, paraboloid..) to set the limit maybe first set within the work space/area or some ratio of existing area... and the external limit would be 1km sq?
    Like • Show 4 Likes4
    Actions
Roland Schwarz Jan 8, 2019 11:55 AM
Mark Correct
Correct Answer
That is a great question.

I was introduced to the concept of untrim in UG(NX). NX remembers the original full untrimmed definition and untrims to that.

SW is kinda arbitrary. 0 equals 100%, unless it means minimal untrim. 100% = 100% unless it means 200%. There's an elegant randomness to the whole thing.
Like • Show 6 Likes6
Actions
- Dan Pihlaja @ Roland Schwarz on Jan 8, 2019 12:09 PM
  Mark Correct
  Correct Answer
  Roland Schwarz wrote:
  
  That is a great question.
  
  I was introduced to the concept of untrim in UG(NX). NX remembers the original full untrimmed definition and untrims to that.
  
  SW is kinda arbitrary. 0 equals 100%, unless it means minimal untrim. 100% = 100% unless it means 200%. There's an elegant randomness to the whole thing.
  
  In this case, there would be no "full untrimmed definition" because this is an imported model. I would assume that this data would not have come through the translation.
  So the question still stands.....why did the first untrim stop where it did? Why not create the cone in the 3rd untrim function?....and why does it go to that particular boundary?
  Like • Show 2 Likes2
  Actions
  - Frederick Law @ Dan Pihlaja on Jan 8, 2019 12:14 PM
    Mark Correct
    Correct Answer
    First one stop at 90 deg?
    Second 180 deg?
    Like • Show 3 Likes3
    Actions
    - Dan Pihlaja @ Frederick Law on Jan 8, 2019 12:30 PM
      Mark Correct
      Correct Answer
      While one side did indeed stop at the next main plane, the other side did not:
      
      Although, this 2nd untrim isn't 90 degrees from the 1st one:
      Like • Show 3 Likes3
      Actions
Matt Lombard Jan 8, 2019 12:30 PM
Unmark Correct
Correct Answer
For some surfaces, it has a meaning. Like a sphere. 100% is easy to visualize for a sphere. A cone is bounded on one end, but on the other it could be infinite, as with other analytical types. But it doesn't really work that way.

You can use Extend where it works better, and sticks to the topological type.

100% has more meaning for a bounded general NURBS surface, where the math only goes so far. Extend can extrapolate, but the curvature gets unpredictable..

So from these experiments, for analytical types, 100% Untrim could/should be predictable, but isn't. For general spline surfaces, 100% untrim appears to be fully predictable.
Attachments
- analytical.SLDPRT.zip231.5 KB
- spline.SLDPRT.zip326.1 KB
Like • Show 7 Likes7
Actions
- Thomas Voetmann @ Matt Lombard on Jan 14, 2019 2:58 AM
  Mark Correct
  Correct Answer
  It is interesting, that if you use untrim/100% one more time on the spline surface, it will extend the surface linear. I can not however figure out how much.
  Like • Show 0 Likes0
  Actions

What does 100% really mean for "Untrim Surface"?

Outcomes

Attachments

Related Content

Recommended Content