Learning Pro/Surface & ISDX

[magneplanar]

james.lynch,

If you read back to the first page, I made the comment that I think pro-e can do G2 to G3 but not full G3. I would contest the notion (if I understand what your comment is?) that ISDX has no greater capability than pro/surface. If that is the case then why have ISDX? They certainly charge a premium for it so it better have functionality not available otherwise. Or did you mean something else?
did the + symbol get you going? (G3+...)

With regard to your comment about pro/surface knowledge, I agree. In fact I would say that using ISDX before reaching a certain capability with pro/surface would be unwise.
...sort of like taking out the race car before you learn to drive...I think they make pro/surface training a prerequisite for the ptc ISDX freeform surface training.

cheers,

M

 

[james.lynch]

james.lynch,

I would contest the notion (if I understand what your comment is?) that ISDX has no greater capability than pro/surface. If that is the case then why have ISDX? They certainly charge a premium for it so it better have functionality not available otherwise. Or did you mean something else?

Hi M.


ISDX has tools which are not available in Pro/Surface for sure, and don't get me wrong, I really like it adn use it day to day and I'm currently asking managment to buy another seat cos we are starting to use it so much. But having said that, there is no surface which can be created in ISDX (in WF2.0) that cannot be done with Pro/Surface.


Can you explain what G2 to G3 means? maybe we have the terminology wrong?

 

[magneplanar]

james.lynch,

I am no "authority" on the subject. However, I was thinking about what you said and I know that curvature is possible in the surface connections made in pro/surface. I do feel that what makes the difference is the ease in which a person can achieve that surface with ISDX. I think ISDX makes the process much simpler and straightforward since the curves and the surfaces are built in the same feature, and can be modified as a group. So I would say if the capability allows the same results with less time or difficulty then it is of greater value.

From the images I have seen of a curvature plot for G2 or G3 surfaces. It seems that the plots for pro/e surfaces look like interpolations of the two. If there is a specific mathematical determination for each G level then I guess there is no such thing as anything in between. Is there such a specific? I am basing my comment on the visual representation and it seems that with the drag handles you can get close to what looks like G3 curvature, and that is the basis for my comment.

cheers,

M

 

[78finn]

Hi there...


Now i'm new to this but why is everyone reffering to Continuity of Curves as G0,G1 & G2 etc?


It is actually reffered to as C0, C1, C2. Also if you are using Pro Engineering Wildfire 3.0 its limitations are C0, C1 & C2. Pro Enginneer currently has no support for C3, Cx etc.


C0 = Two surfaces have C0 continuity if they meet along a common edge


C1= Two surfaces have C1 continuity if they are are tangent along a common edge.


C2 = Two surfaces have C2 continuity if there curvature across the common edge is continuous or 'smooth'. This makes the transition between surfaces invisible to the eye.


C3 & C infinity, are only used in programs like Rhino and are areally of little use to anyone unless invovled in Areonotical engineering.



< =text/>_popupControl();

 

[magneplanar]

78finn,

you are correct in part...there is definition for G and C. Each has specific characteristics. "G" is sometimes referred to as geometric continuity. Both are often intermixed in their usage although they are not exactly equal. The link I sent earlier in this topic showed images with G connections.

<h3>Geometric and Parametric Continuity </h3>


Geometric Continuity
<ul>[*]G₀: curves are joined
[*]G₁: first derivatives are proportional at the join point

The curve tangents thus have the same direction, but not necessarily the same magnitude.

i.e., C₁'(1) = (a,b,c) and C₂'(0) = (k*a, k*b, k*c).
[*]G₂: first and second derivatives are proportional at join point [/list]


Parametric Continuity
<ul>[*]C₀: curves are joined
[*]C₁: first derivatives equal
[*]C₂: first and second derivatives are equal

If t is taken to be time, this implies that the acceleration is continuous.
[*]C_n: nth derivatives are equal [/list]


As their names imply, geometric continuity requires the geometry to
be continuous, while parametric continuity requires that the underlying
parameterization be continuous as well.


Parametric continuity of order n implies geometric continuity of order n, but not vice-versa.
take a look at these links:

[url]http://www.cs.mtu.edu/~shene/COURSES/cs3621/NOTES/curves/con tinuity.html[/url]

[url]http://www.cs.toronto.edu/~faisal/teaching/notes/csc418/fais al/topics11.html[/url]

cheers,

M


Edited by: magneplanar

 

[pjw]

Curvature continuity isn't all that involved. In a nutshell it is only the second derivative of position where all points edges of separate surfaces meet. Basically, it forces the same curvature between two different surfaces a set distance from their meeting point. The joint becomes 'seamless' or 'invisible' to the naked eye. Higher orders simply mean that the curvature is forced to be the same further alon from the seam. It is used widely in industrial design and the automotive sector.


Incidentally, does anyone know the best way to put a curvature continuous fillet on two perpendicular solid surfaces in Pro/E? It's not as easy as it sounds!

 

[james.lynch]

hmm G2 rad on two planar perpendicular surfaces would not be easy (or perhaps desirable??) I don't think it could be done without having a slight inflection in the surface.


Because it's a planar surface, and because C=1/R and for a flat surface R is infinity, therefore,C = 1/infinity = 0


This leads to problems..

 

[design-engine]

Pro/E is a little far off from being able to take seriously from being a
full g2 modeler. Much less g3. For example. You can
manipulate internal CV's (control vertices) of the curve with g1
curves. When you force g2 you loose that manipulation function.
Secondly until PTC grants users the ability to force one CV a snap function to
another CV in another feature (or inside the current ISDX feature) users can't
do full direct model techniques like auto designers do with Alias Studio.
Thirdly, until PTC offers users the ability to convert a 3 degree curve to a 4,
5, 6 or 7 degree curve and back again we will not even have the vocabulary to
compare to an Alias Studio. ISDX in WF 4.0 is making one more step
towards that direction however I think the folks developing ISDX need a 2 day
crash course in Alias A-class surfacing.<u1></u1>



Technically Pro/E can get close to g3 but with un-wielding control.



I could post some examples of a 3 degree curve converting to a 4 degree
curve with respect to Alias Studio if anyone cares. modeling techniques
become an open door for proving form and users are not just limited to doing
everything with four part boundaries. I had this discussion
with the new product line manager of Surfacing tools for PTC. Very nice fellow
however he was just being polite and brushed me off. They just had
their technical committee meeting too. I need to be more persistent and
try to influence WF5 release. I don't think I should be getting in this kind of detail in a discussion titled "learning ISDX" either.



Maybe a little harsh because I just got in an argument with the local cop
in my nationhood for giving me a parking ticket in a clearly ambiguous
situation.




Edited by: design-engine

 

[Jacoolo]

Bart you mentioned very intersting things considering difference between Pro\e and more advanced surf application. I am very curious of real examples of these differences. Btw - is Alias also a feature based parametric tool?

p.s Police, yeah, can take off the smile from your face for the whole day. Well 3 weeks ago I was going to fast on the road - 30km\h more than allowed. I was catched by police. I kept my mouth shut, I did not go into disscusion, I agreed it was my fault to go to fast and in the end -... they let me go with no problems. So it is possible to meet nice policeman))

 

[design-engine]

you make me feel better but I pay over 2 thousand per year US in parking tickets. I am making a serious change to not park improperly to reduce the amount and where I parked was fine.

 

[Jacoolo]

I thought that wives and girlfriens are expensive - but I say: no way. Parkings in the USA are much more... ))

 

[78finn]

< =text/>_popupControl();
Ok peeps....Feel as if we might have strayed from the original point of this discusion

which was to help us umble learners to get to grips with ISDX...


I have a question though. I have been using Pro Surface for a few weeks now...and am getting along quite nicely there. I like the fact that it it dimension driven and can be heavily constrained....I work as an industrial designer, so having total and accurate control over a surface is a must for me...nothing I do is purely vsiual...mostly injection moulded plastic components.


So can anyone tell me - in laymens terms would be best - how you accurately work with ISDX? I assume by difining datum poits curves etc etc....but do you make them prior or during. Again....I wouod have thhought both.


It seems to me that ISDX is largely visual based....?? Am I right? Or horribly wrong?? Love the way it works though....like 3D Max on steroids!

 

[pjw]

Use skeleton planes and datum curves and points prior to creating your ISDX features. I think the whole point of having ISDX is the fact that geometry is created largely freeform - i.e. you needn't worry about dimensions. All well and good for the industrial designer who is more interested in form than precision. High level surfacing is near to impossible to check dimensionally. You can make datum features during the creation of a style surface but I think you will have better control over the surface if you make the datum features first.


Hope this helps,


Phil