CAD/CAM discussion forum > 3D CAD/CAM > How do you bend this?

How do you bend this?

    
  Subscribe Topic

Rank: 7Rank: 7Rank: 7

Paul

Moderator

posts: 333

Registered: 2011-9-17

Message 1 of 10

 How do you bend this?
18-11-2009 07:04 . pm   |   View his/her posts only
Is a bend a morph? OK, it is but can VX do the bend?

The attached model shows the starting material and the curve list where I'd like it to end.
I could have optionally swept and parallel notched then straightened to see what the initial notch removal looks like.

I would expect to the see the notches close up during the bend.
I have tried morphing cannot see how to do it.

Is it possible and how?

Cheers

Rank: 1

Mike

Newbie

posts: 0

Registered: 2002-8-28

Message 2 of 10

19-11-2009 02:11 . pm   |   View his/her posts only
Paul,

"Morph curve to curve" will do a pretty good job on this piece if you use an influence around 2000. You should also be able to sweep the shape around the bend and also do a circular pattern on the extrude cut feature.

Rank: 1

smd

Newbie

posts: 0

Registered: 2011-12-31

Message 3 of 10

19-11-2009 03:04 . pm   |   View his/her posts only
You can morph curve to curve to bend this. There's a ton of geometry with those cut outs. It went fairly quick without them, just bending the long rectangle, but with them it was painful. Put a sketch the lenght of the rect. on the side that would coincide with your curve list , pick morph crv to crv select shape put pick filter to curve, select skecth for first curve, then your curve list, and pay attention to which way the arrows are goin.....play with the bend translate buttons as well as the influence and watch the preview. Try it with the basic shape first, then with the cut outs. i think this file is too big for this forum ...oh wait, i've got your email, i'll send thata way. Scott


Oh and what ever he said... And Mike is this where the reduce file size option will come in on the morph command in V14.2? Or was I makin that up in my head?

Rank: 1

smd

Newbie

posts: 0

Registered: 2011-12-31

Message 4 of 10

20-11-2009 09:23 . am   |   View his/her posts only
Here's another that is simpler geometry morpfed in two directions....was anyone able to open that last one?

Rank: 1

Mike

Newbie

posts: 0

Registered: 2002-8-28

Message 5 of 10

20-11-2009 11:23 . am   |   View his/her posts only
I can open both of your files.
Be careful not to minimize the Morph influence too much. You used 900. See the far end (left). It is not square like the original extrude.

Rank: 7Rank: 7Rank: 7

Paul

Moderator

posts: 333

Registered: 2011-9-17

Message 6 of 10

20-11-2009 01:41 . pm   |   View his/her posts only
Hi Mike,
how would a user ever figure what the Influence number values mean? What clues are there that 900 or 2000 is a reasonable value.
The default is 20! Why not work in perentages? We can all understand and relate to that. Seems to me there must be a 100% value at which point influence is meaningless.....

Scott is correct when he says it can take a long time to compute. This is a LOT different to the on screen 'instant' morphing in free form modellers.
Can you please explain why the time issue comparitively? What we can do to reduce that etc.

Another point is the redefining a morph doesn't give too many clues as to the existing picks (apart from geometry)- especially when it takes a while to preview. All highlights are the same colour - just blur of yellow.

Apart from that I think the results are OK.
A lot of parts are actually bent from straight so morphing needs to be able to do this job representative of an actual bend.

FYI. I have actually built this part with much smaller cutouts. In reality the bending occurs in the small cross section zone and you end up with a series of flats. A bending model would consider this and represent the material deformation.

In determining the cutout requirement I Inquired the respective length if inner and outer curves. What is curious is that the curve length value is not available as a dimension for use in an equation. IMO any 'Inquirable' value should be accessible rather than a screen transient. Or have I missed how to do this?

Cheers

Cheers

Rank: 1

Mike

Newbie

posts: 0

Registered: 2002-8-28

Message 7 of 10

20-11-2009 02:35 . pm   |   View his/her posts only
Simple answer is I started with 20 and it looked like crap. We have to put some value in there. I don't know all the heuristics that go into writing a command like this but I do know it is not like the sheet metal command that actually bends and deforms metal. When I looked closely at the model I noticed that the farther away from the common end, the part was bent. Increasing the value helped to straighten it out.
From the on-line HELP - Influence (I) Specify radius of influence. The smaller the number the more localized the change. A larger radius value will influence more of the part. This is alot better than Version 12.93 Influence (I) - Specify radius of influence.
Inquire curve length does have a variable option.

Rank: 1

Randy

Newbie

posts: 0

Registered: 2006-5-18

Message 8 of 10

20-11-2009 02:44 . pm   |   View his/her posts only
Paul,
While waiting for Mikes reply to your questions I decided to play around with Scott's morph. It appears that the "Influence" value only needs to be as large as a radial distance from the bending curve that would envelope the area to be bent.
In this case I used an Influence value of only 200, a Rigid value of 100%, Bulge and Slope values of 0% with good results.

Rank: 1

Mike

Newbie

posts: 0

Registered: 2002-8-28

Message 9 of 10

20-11-2009 03:41 . pm   |   View his/her posts only
In speaking with the developers, morph may not be a precise solution. When you bend this part you want the gaps to close up but in reality even the teeth are collapsing by the same ratio. I think you will have to sweep the part w/o the grooves; then do a circular pattern.

Rank: 7Rank: 7Rank: 7

Paul

Moderator

posts: 333

Registered: 2011-9-17

Message 10 of 10

21-11-2009 02:21 . am   |   View his/her posts only
Hi Mike,
my apology to the VX team about the selection during refine. In 14.+ it is all good.

Wassmer is correct as far as getting it sorted. One has to come to grips with the controls BEFORE applying to a complex model.
Take away the complexity and it is quite fast.
My future protocol would be to use a simple shape to develop the morph regime then used the values as a guide for the more complex shape.
I think there is a place for some sort of graphical guide example set (similar to fillet scenarios) showing the results of various morph settings with each command.
The issue is understanding how it works.

Scotts' locket example is excellent as it shows how compound curvature can be added to the part after the majority of the modelling. Chris showed that as well but the model is explanatory compared to an example picture.

I am just starting to get my head around morhing and really appreciate the input from all of you.

Will try and hunt out that inquire variable.

Cheers
See also
X