Round won't work, but should

[hallstevenson]

Can anyone explain why I can't make the .05" round to be .06" ?? I did the "hard" round first and it obviously contributes to the problem.

Call this part a "concept". My real part has dozens of features that mate/intersect similarly, all requiring rounds blended between them.

[Anonymous]

maybe someone has a better way

[Anonymous]

Edit the "Hard" fillet. At the vertex where the 5 edges come together
add a setback and set the value for each edge to .05. do this for both
verticies. Then the .06 fillet will go in no problem.

-Kevin

diemaker wrote:
> maybe someone has a better way

[Anonymous]

excellent... glad I took a look at this thread.

[hallstevenson]

> At the vertex where the 5 edges come together add a setback
> and set the value for each edge to .05. do this for both verticies.

Sorry if the .05 radius confused things. I want (need ??) all radii to be .06. I only made the one .05 to get *something* in there that worked.

Can I ask you why it fails ?? From my experience, what I call the "hard" fillet has to be done in (1) feature, selecting the (4) verticals and (3) horizontals, because some of the edges have mat'l removed while others have it added. If you don't combine the fillet into one feature, it won't blend later.

[Anonymous]

This was one of the shortcomings of the ACIS kernel (Shape Manager now). If
you could post a ACIS version of the file, I would like to try it on a
Parasolid modeler to see if the same problem occurs.

Ken
wrote in message news:4831266@discussion.autodesk.com...
> At the vertex where the 5 edges come together add a setback
> and set the value for each edge to .05. do this for both verticies.

Sorry if the .05 radius confused things. I want (need ??) all radii to be
.06. I only made the one .05 to get *something* in there that worked.

Can I ask you why it fails ?? From my experience, what I call the "hard"
fillet has to be done in (1) feature, selecting the (4) verticals and (3)
horizontals, because some of the edges have mat'l removed while others have
it added. If you don't combine the fillet into one feature, it won't blend
later.

[hallstevenson]

Who all uses ACIS and who uses Parasolid ??

[Anonymous]

It is not an ASM or Parisolids issue. Inventor CAN make the blend you
are looking for.

Go back to the model you posted.
Edit the first fillet.
In the dialog select the last tab called setback. add the two vertices
and set all the edges to .06 the same as the radius
Edit the second fillet. Add the second edge that you wanted and changed
the .05 radius to .06. the fillet will finish just fine.

-Kevin

hallstevenson wrote:
> Who all uses ACIS and who uses Parasolid ??

[Anonymous]

hall: kevin said use a .050 setback, but i think he meant .060 cause that gives a smooth r.060 all around the corner and allows the r.060 around the edge.

IV's radius has come along way from a couple of years ago.

[firefly]

I have found why you can put another .06 fillet
what you are doing wrong is the fillets are overlapping and the program doesn't like it so i wont allow you to do it. if i were you and the .06" fillet was required than i would do is make the second extrusions smaller by .01" on both sides

[Anonymous]

The confusion may be the terminology of set back.
Checking IV help file, setback is a distance from the vertex
and does not change the radius.
Have attached two images from IV help file explaining
the setback.

my two pence.

Wolfe

"Kevin Schneider (Autodesk inc.)" wrote in message
news:4831363@discussion.autodesk.com...
It is not an ASM or Parisolids issue. Inventor CAN make the blend you
are looking for.

Go back to the model you posted.
Edit the first fillet.
In the dialog select the last tab called setback. add the two vertices
and set all the edges to .06 the same as the radius
Edit the second fillet. Add the second edge that you wanted and changed
the .05 radius to .06. the fillet will finish just fine.

-Kevin

hallstevenson wrote:
> Who all uses ACIS and who uses Parasolid ??

[donb]

Thanks, Kevin. There's got to be a lot of users, myself included, having the same problem. Your answer fills that vital gap of understanding between what the help file shows and what some (many?) of us need to know, to get the results we need.

The use of setback to solve this kind of filleting problem wasn't covered either in formal training I took (levels 1 and 2) last fall, nor in 3rd party manuals I have. As a result, I have had some real problems providing desired fillets on several models. Now I see that most or all could have been done by adjusting setback.

It would be nice if an example like this one could get wider distribution than to those following this discussion group.

Don Bruhns

[xavierl]

well said.

it would be nice if the online help discussed each command ie filllet and systematically went through each feature.
regards
Frans X Liebenberg

[hallstevenson]

Is the setback "tightening" up how far the blends extend ?? If you increase the width of the lower, larger block, to get the corner/edge away from the blend, .06 works all around. No setbacks necessary.

I see what the problem is and why IV "fails" (without further massaging). The attached image shows what/where I think the problem is. Am I correct ??

If there are any plans to make rounds better or "more robust", maybe this can be improved upon. The same round works fine in ProE...

[Anonymous]

the extra r.060s that extend where they shouldn't might be as designed... can proE make the fillet that IV defaults to? Or does it only fillet like you wanted it THIS time.

Or it might be IV just can't figuere what to fillet and what to tangent. So they fillet everything and provide a way to adjust. It's obviuos what you want to do, but if you had to describe it exactly, in writing, without pointing, so a 10 year old could understand...it would be very hard.

[hallstevenson]

ProE builds the round using .06, with no add'l steps necessary. You still have to do the "hard" round first, exactly the same as in IV but then then round that was failing in IV builds fine, using all default settings also.

[Larry Caldwell]

... by Jiminy!
~Larry