Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Splitting a curve 3
- Thread starter bfleck
- Start date
-
1
- #2
NX version please.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
- Thread starter
- #3
I guess the version won't matter....just some minor dialog changes from NX3 to NX6.
The answer is yes you can (more or less). Look on the Edit Curve toolbar for a command called Divide Curve. Now, you cannot use a surface, but what you can do is create a Point (associative or not) at the intersection of the surface and the curve, then use the Point as the dividing object (rather than the surface). A Plane is fine, without any extra geometry creation.
When you first bring up the Divide Curve command (if using NX5 or higher), you may have to set the pulldown at the top of the dialog to be the correct type of Divide to be performed. It's fairly easy to figure it out, I think.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
The answer is yes you can (more or less). Look on the Edit Curve toolbar for a command called Divide Curve. Now, you cannot use a surface, but what you can do is create a Point (associative or not) at the intersection of the surface and the curve, then use the Point as the dividing object (rather than the surface). A Plane is fine, without any extra geometry creation.
When you first bring up the Divide Curve command (if using NX5 or higher), you may have to set the pulldown at the top of the dialog to be the correct type of Divide to be performed. It's fairly easy to figure it out, I think.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
- Thread starter
- #5
- Thread starter
- #6
-
1
- #8
Trim curve can create an associative feature. The trick is start with a base curve and then trim each section separately. The good new is that it will trim and extend so you can get something back for having to put up with the extra selection steps.
Cheers
Hudson
Cheers
Hudson
bfleck,
Had I known you were wanting associativity, I might have responded differently yesterday. If you don't include all pertinent information in your initial query, then others may potentially waste time responding to what amounts to only part of an issue.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
Had I known you were wanting associativity, I might have responded differently yesterday. If you don't include all pertinent information in your initial query, then others may potentially waste time responding to what amounts to only part of an issue.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
- Thread starter
- #10
JohnRBaker
Mechanical
"Why would a designer want features that are not associative."
You'd be amazed at some of the arguments that have been made in support of the opposite view-point ;-)
John R. Baker, P.E.
Product 'Evangelist'
NX Design
Siemens PLM Software Inc.
Cypress, CA
To an Engineer, the glass is twice as big as it needs to be.
You'd be amazed at some of the arguments that have been made in support of the opposite view-point ;-)
John R. Baker, P.E.
Product 'Evangelist'
NX Design
Siemens PLM Software Inc.
Cypress, CA
To an Engineer, the glass is twice as big as it needs to be.
bfleck,
Good question, but I feel you're asking me to assume quite a bit.
Now, I'm not trying to pick on you or tell you how to post a question. I'm just trying to help you get your questions answered as quickly as possible and as correct as possible the first time. I am also not speaking for everyone else here, just myself.
You were working with a curve, which could mean you were doing one of about a thousand different tasks or workflows in NX and that is all I (or anyone else) knew at the start of this thread. There was no indication that associativity was desired or not desired or what the curve might be used for downstream. Based on that, how are we to know what you may or may not want for a response or suggestion? My crystal ball has never been THAT powerful. ;-)
Maybe time is the only thing that matters to a particular designer and they could care less about associativity because they're creating concept models or something that will never need to be changed. Maybe they are creating models for one-off parts that will never be made again. Who knows? I sure don't, but your statement above implies that I should assume they are wanting full associativity and I should respond with that in mind rather than make SURE my response will fit into the designer's workflow or needs.
I happen to use Divide Curve all the time for building non-associative splines used to create slab surfaces with Thru Curve Mesh. Those slab surfaces are never going to change because they affect the aesthetics more than other surfaces might. However, at a later time, I may need to change the blends or transitional surfaces that fit between the mesh surfaces to avoid a durability issue, which is inherent to our product, so those surfaces I might create with associativity or parametrics. It just depends on how much time I have and how necessary I feel it is.
When offering advice to those in need, I feel that it's not a good idea for me to assume that NX is always being used the same way and for the same purposes. That would fall in line with me assuming that everyone is using the exact same version of NX when the reality of it is that there are still people out there not even using the NX series yet. By providing clear and accurate information from the beginning, you are helping to ensure that the advice you are getting in return is as sound as possible the first time rather than the second, third or fourth time....as well as saving time for the people who are responding to your inquiry by eliminating questions that may come to mind when trying to figure out a solution.
Mr. Baker,
Please, if I haven't done a good job of explaining this, feel free to offer specific examples that you have seen, as I expect you have a broader experience with seeing different workflows/usage of NX over the years.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
Good question, but I feel you're asking me to assume quite a bit.
Now, I'm not trying to pick on you or tell you how to post a question. I'm just trying to help you get your questions answered as quickly as possible and as correct as possible the first time. I am also not speaking for everyone else here, just myself.
You were working with a curve, which could mean you were doing one of about a thousand different tasks or workflows in NX and that is all I (or anyone else) knew at the start of this thread. There was no indication that associativity was desired or not desired or what the curve might be used for downstream. Based on that, how are we to know what you may or may not want for a response or suggestion? My crystal ball has never been THAT powerful. ;-)
Maybe time is the only thing that matters to a particular designer and they could care less about associativity because they're creating concept models or something that will never need to be changed. Maybe they are creating models for one-off parts that will never be made again. Who knows? I sure don't, but your statement above implies that I should assume they are wanting full associativity and I should respond with that in mind rather than make SURE my response will fit into the designer's workflow or needs.
I happen to use Divide Curve all the time for building non-associative splines used to create slab surfaces with Thru Curve Mesh. Those slab surfaces are never going to change because they affect the aesthetics more than other surfaces might. However, at a later time, I may need to change the blends or transitional surfaces that fit between the mesh surfaces to avoid a durability issue, which is inherent to our product, so those surfaces I might create with associativity or parametrics. It just depends on how much time I have and how necessary I feel it is.
When offering advice to those in need, I feel that it's not a good idea for me to assume that NX is always being used the same way and for the same purposes. That would fall in line with me assuming that everyone is using the exact same version of NX when the reality of it is that there are still people out there not even using the NX series yet. By providing clear and accurate information from the beginning, you are helping to ensure that the advice you are getting in return is as sound as possible the first time rather than the second, third or fourth time....as well as saving time for the people who are responding to your inquiry by eliminating questions that may come to mind when trying to figure out a solution.
Mr. Baker,
Please, if I haven't done a good job of explaining this, feel free to offer specific examples that you have seen, as I expect you have a broader experience with seeing different workflows/usage of NX over the years.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
Tim,
For what it's worth I wouldn't have guesses either and wouldn't have bothered to add to your earlier response thinking it sufficient prior to the additional request being posted.
Bfleck,
The short answer is that there are times when associative curves are desirable and for those reasons NX should and in the main does support methods to achieve those results.
In the main however curve creation has a longer history of being non-associative and many users are happier to work in that manner.
Technically any model has to be based on something before anything becomes associative to basic features. Depending on what you intend to use the curves for sometimes the associativity may be of limited use or relevance to your design. Curve creation being most commonly basic to building surfaces most users find that it is somewhat easier and cleaner to limit the amount of curve associativity that they employ in building those models. In that way some if not most feel that they don't need associativity for working in curves compared with the benefits one finds in working with feature based solid models. Thus the natural assumption for many is not to require associative curve trimming.
In your case if you need to preserve the associativity then you have your answer. In my experience most users don't regularly insist on making curve trims associative.
Cheers
Hudson
For what it's worth I wouldn't have guesses either and wouldn't have bothered to add to your earlier response thinking it sufficient prior to the additional request being posted.
Bfleck,
The short answer is that there are times when associative curves are desirable and for those reasons NX should and in the main does support methods to achieve those results.
In the main however curve creation has a longer history of being non-associative and many users are happier to work in that manner.
Technically any model has to be based on something before anything becomes associative to basic features. Depending on what you intend to use the curves for sometimes the associativity may be of limited use or relevance to your design. Curve creation being most commonly basic to building surfaces most users find that it is somewhat easier and cleaner to limit the amount of curve associativity that they employ in building those models. In that way some if not most feel that they don't need associativity for working in curves compared with the benefits one finds in working with feature based solid models. Thus the natural assumption for many is not to require associative curve trimming.
In your case if you need to preserve the associativity then you have your answer. In my experience most users don't regularly insist on making curve trims associative.
Cheers
Hudson
hudson,
I completely understand and appreciate all of your posts, this one being no different. I tend not to include methods outside of my typical workflows, as I would be concerned about trying to pass along detailed, step by step processes and missing a vital part.
I never would have considered associative Trim Curves, just due to my experiences with them as well as it possibly becoming difficult to reassociate those types of edits to completely different geometry later on in the lifecycle. Not everyone edits a model in the same manner...that is another consideration for us to think about when responding to an inquiry as well.
My high school teachers would never believe that I'm actually taking something in my life so seriously. ;-)
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
I completely understand and appreciate all of your posts, this one being no different. I tend not to include methods outside of my typical workflows, as I would be concerned about trying to pass along detailed, step by step processes and missing a vital part.
I never would have considered associative Trim Curves, just due to my experiences with them as well as it possibly becoming difficult to reassociate those types of edits to completely different geometry later on in the lifecycle. Not everyone edits a model in the same manner...that is another consideration for us to think about when responding to an inquiry as well.
My high school teachers would never believe that I'm actually taking something in my life so seriously. ;-)
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
Tim,
Thanks for your post. I got the sense that you might have fallen foul of a typo in the first sentence of my previous post; it should read "I wouldn't have guessed either". I wasn't suggesting you were posting based on blind guesswork or any such thing.![[wink]](/data/assets/smilies/wink.gif "[wink] [wink]")
90% of posters on the forum don't give enough information about their question. Maybe on reflection we could add that bfleck might do better to have used a sketch. I can't say if that's true because I simply don't have enough to go on. Many's the time we've argued the relative merits of a particular feature only to find the poster coming back with a clarification that points to a completely different subject.
Getting the question right is half the battle sometimes it just comes down to jargon and can't be helped. So we can't be too hard on people asking questions; if they knew how to get the question right they probably wouldn't have asked, instead they'd probably have found it in the manual without our help.
Sometimes I am keen to venture outside my typical working methods because I do find that I learn from that. Even if I learn why people think in ways I would never consider or just plain disagree with. I do however try to test anything that I post before reeling off the steps. If I can't I try to say I'm not sure and admit that I'm wrong when necessary. It is when I'm wrong that I'm able to learn things.
Truth be told when it comes to associativity it is often hard enough trying to figure out how somebody else built their model using features without having to analyse curve dependencies as well. I completely appreciate what you're saying in that regard.
I think we probably work the same way in such things. But you know we may say so here at some risk of starting yet another argument with some users who find this flies in the face of their strong opinions to the contrary.
Cheers
Hudson
Thanks for your post. I got the sense that you might have fallen foul of a typo in the first sentence of my previous post; it should read "I wouldn't have guessed either". I wasn't suggesting you were posting based on blind guesswork or any such thing.
90% of posters on the forum don't give enough information about their question. Maybe on reflection we could add that bfleck might do better to have used a sketch. I can't say if that's true because I simply don't have enough to go on. Many's the time we've argued the relative merits of a particular feature only to find the poster coming back with a clarification that points to a completely different subject.
Getting the question right is half the battle sometimes it just comes down to jargon and can't be helped. So we can't be too hard on people asking questions; if they knew how to get the question right they probably wouldn't have asked, instead they'd probably have found it in the manual without our help.
Sometimes I am keen to venture outside my typical working methods because I do find that I learn from that. Even if I learn why people think in ways I would never consider or just plain disagree with. I do however try to test anything that I post before reeling off the steps. If I can't I try to say I'm not sure and admit that I'm wrong when necessary. It is when I'm wrong that I'm able to learn things.
Truth be told when it comes to associativity it is often hard enough trying to figure out how somebody else built their model using features without having to analyse curve dependencies as well. I completely appreciate what you're saying in that regard.
I think we probably work the same way in such things. But you know we may say so here at some risk of starting yet another argument with some users who find this flies in the face of their strong opinions to the contrary.
Cheers
Hudson
hudson,
Oh no, I understood what you were saying in your previous post. No misunderstandings whatsoever here.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
Oh no, I understood what you were saying in your previous post. No misunderstandings whatsoever here.
Tim Flater
Senior Designer
Enkei America, Inc.
Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.
- Thread starter
- #17
It also shows me that people do not design how I would design and I had a lapse of thought when I posted the original question. I always want associative features, because I believe that is the beauty of parametric modeling. Certainly someone may want a quick representation of a part, but what happens if they need to change it. Spending the time to be associative can save down the road.
-
1
- #18
Associativity is a double edged sword; with it you can build a time saving wonder part or a hopelessly tangled mess. Often you are not sure which you have until the change orders start coming in.
It is smart to build associative parts, it is wise to know when not to.
It is smart to build associative parts, it is wise to know when not to.
- Thread starter
- #19
bfleck,
Yours above is not a silly comment and why many organisations recognise doing things in different ways at different phases of a project. To begin with you might model a layout chaotically based on the very idea that only you need to work with the data so it isn't expected to be maintainable by others. This enables an initial response to pressures to achieve proof of concept. Good organisations often go back over their designs to clean them up for modelling in ways that are maintainable for the design to accommodate tooling changes or support future variants.
The difference between the earlier and later stage models is the intention to be maintainable. Sometimes you can model that way from the start. Models that are simple enough in that respect are usually below a level of complexity that challenges NX in any case.
Associativity is at its best when geared to maintainability. But it can also be a weakness if it involves a level of dependency that requires so much in depth analysis that it takes too long; perhaps because you find yourself stuck in the process of updating a great many dependent features when performing what you thought was a minor manipulation of the data. As many of us will attest there are ways to avoid modelling oneself into a corner by insisting on maintaining complete associativity when it may not be really necessary to do so. A lot of features can be maintained relative to one another by taking measurements as required or edited by selecting alternate geometry when needed so the number of requirements for internal interdependencies can be relaxed without necessarily affecting the design.
Cheers
Hudson
Yours above is not a silly comment and why many organisations recognise doing things in different ways at different phases of a project. To begin with you might model a layout chaotically based on the very idea that only you need to work with the data so it isn't expected to be maintainable by others. This enables an initial response to pressures to achieve proof of concept. Good organisations often go back over their designs to clean them up for modelling in ways that are maintainable for the design to accommodate tooling changes or support future variants.
The difference between the earlier and later stage models is the intention to be maintainable. Sometimes you can model that way from the start. Models that are simple enough in that respect are usually below a level of complexity that challenges NX in any case.
Associativity is at its best when geared to maintainability. But it can also be a weakness if it involves a level of dependency that requires so much in depth analysis that it takes too long; perhaps because you find yourself stuck in the process of updating a great many dependent features when performing what you thought was a minor manipulation of the data. As many of us will attest there are ways to avoid modelling oneself into a corner by insisting on maintaining complete associativity when it may not be really necessary to do so. A lot of features can be maintained relative to one another by taking measurements as required or edited by selecting alternate geometry when needed so the number of requirements for internal interdependencies can be relaxed without necessarily affecting the design.
Cheers
Hudson
- Status
