Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

mirror, then rotate?

651 views
Skip to first unread message

Scott Coburn

unread,
May 16, 2001, 4:37:46 PM5/16/01
to

Hello,

I am creating a part which is to be a disk with some extruded features
on each of the two faces of the disk. The features on each side are
identical, but are rotated 90 degrees relative to the set on the
other side.

I have made "half" of the part (a half-thickness disk plus the features
on one of its faces). I can mirror this part to get the disk to full
thickness and the features placed on the back face, but these features
now need to be rotated 90 degrees.

I can think of a few of ways to do this:

1) create an assembly file with the two halves constrained the way
I want them. (Giving me two files instead of a nice neat "one",
and the resulting "part" has a seam showing where the two disks
mate.)

2) make the disk full thickness and redraw or copy all of the features
to the other side. (A real pain with lots of features.)

3) make the half disk with features on one side a design element
and attach them to ? (a thin disk?). Again, this looks like
a two file solution.

So, there must be a "right" way to do this.

Scott

--
------------------------------------------------------------------------
Scott Coburn sc...@bnl.gov
Brookhaven National Laboratory
Building 510B tel 631 344 7110
Upton NY 11973-5000 fax 631 344 2739
------------------------------------------------------------------------

Quinn Zander

unread,
May 16, 2001, 4:51:05 PM5/16/01
to
Scott,

After creating the assembly, open a new part file (get out of the sketch)
and insert the assembly as a derived part... select not to show edges... you
will end up with a "seamless" part.

I think this is the simplest method. DesEls would work, but you start
limiting yourself if the extruded features in themselves need to be changed
to any great extent.

QBZ

Scott Coburn

unread,
May 16, 2001, 5:49:31 PM5/16/01
to

Quinn Zander wrote:
>
> After creating the assembly, open a new part file (get out of the sketch)
> and insert the assembly as a derived part... select not to show edges... you
> will end up with a "seamless" part.

OK, that looks right, but now I have three files?!

Next "problem":

There is a pattern of counterbored holes on each side of this part.
Like the features, the patterns are identical but rotated 90 degrees.
So, I can put the hole pattern into my "half part" with "Thru All"
attribute but when I assemble the two parts and then create a derived
part the holes do not go all the way through. (They go all the
way through the "half part", but (as expected) they don't continue
through the other "half part".)

I can add the hole pattern to my derived part, but then how do I
get it to the other side, rotated by 90 degrees? (There is more
than one hole pattern on each side...)

If I was to put myself into the shoes of the guy who had to actually
code what I want to do I would have to admit that it is a bit of a tall
order. But, as long as I am wishing...

Drew Fulford

unread,
May 16, 2001, 9:26:28 PM5/16/01
to
Scott.

Try this.

Draw your whole disk. Add your 20 odd features to your one side. then
in the browser highlite the 20 odd features and rmb choose copy them
all in one shot. Flip the disk over and highlite the opposite face.
Then RMB and choose paste. Click the parameters setting to "Dependant"
and then plug a 90 deg (or what look right) in the angle box.

Is that what you want? Or did i miss a mirror in there? Not sure since
you said a desel may work which is not mirrored but copied just like
this approach.

- drew

On Wed, 16 May 2001 13:37:46 -0700, Scott Coburn <sc...@bnl.gov>
wrote:

Drew Fulford, B.A.Sc. Systems Engineer
Solid Caddgroup Inc, Burlington, Ontario
Phone: (905) 331-9670 Fax: (905) 331-7280
Check out my MDT & Inventor FAQ @ http://www.mymcad.com/

Scott Coburn

unread,
May 21, 2001, 4:06:17 PM5/21/01
to

Josh,

This worked _perfectly_, doing exactly what I wanted!

Great program. Still needs osnaps though.

Thanks,

Scott


jtrain wrote:
>
> Isn't there a line of symetry at 45 degrees rotation from one of the features?
> I would model half of the disk (180 degrees)with the flat side being along
> this line of symetry, place the features on one side, then do a circular array
> of two instances at 180 degrees. I will post a file to CF as Disk Array.
> Josh Trainor

0 new messages