non-cutting moves....is this a reported glitch? 1

[capnhook]

First post here, UG CAD/CAM user since the early 1990's at GM, exiting UGNX at version 4.x, and returning now again to NX 7.0.1.7..........after 1.5 years immersed in the CATIA CAM world. What a long, strange trip it's been..

Is there anything to look forward to, in regards to improved non-cutting moves being implemented into NX 7.5 and beyond? After being spoiled with all the versatility that CATIA CAM has with its nearly infinite variations of control over non-cutting moves, I find myself today already stumped with the erratic output NX gave me for a relatively simple off-plane arc retract.

Working a 5-axis cut down into the part in three multiple levels, the initial and final retracts generate correctly, but the retract on the middle cut level arcs off in the totally opposite direction. I have attached a .jpeg file of what NX generates, and the fairly simple variable set that I used to get this erroneous output.

Has gtac had any reports on this behavior, or should I order padding for the rubber room I am going to have to live in from now on in order to be able to watch this NX CAM like a hawk?

It's great to be back,
capnhook


Proud Member of the Reality-Based Community..

 

[wmalan]

What type of operation is this? Is it a closed boundary or open? Also it's hard to tell in your pic but - that wall looks like some kind of B-Spline surface that bows out at the center. I know you may have your reasons but... that 5.000" linear extension value is pretty large coming off the face. It may be having problems projecting it. Just for kicks reduce the linear extension to a small value and see what happens. Would be interesting to see...

--
Bill

 

[cadcamnx]

friend nxcam changed much your exit strategy tools
was not like now moves non cuting this more complete understand their difficulty
GTAC but will not help you also work with 5 axis and at first it was hard
but I soon adapted you recommend a course of upgrading NX7 because at nx 07/05
has not changed anything yet

or if you want me to pass the problem on my e-mail
We can do this upgrade online on skype
andrenxcam@yahoo.com.br

hug !!!!!!


Andre

specialist and multi axis
development of IS & V
and postprocessors
member committee nxcam plm world
skype:andrenxcam7

 

[capnhook]

This is a variable-contour, surface area drive operation....tool axis normal to drive.

Taking a closer look, I reduced the "Arc-Parallel to Tool Axis > Swing Angle" to -30 degrees, and shortened the "Linear Extension" to 0.5" (which really should have no bearing, since it occurs *after* the arc retract movement). I also upped the number of cut layers to five, from three, for illustration.

If you look at this next attached file, the retract moves are even more screwed up. Counting from top to bottom, cutlayer1 swings in the wrong direction, cutlayers2&3 retract properly, cutlayer4 swings in the wrong direction, and cutlayer5 retracts properly!

3 correct moves out of 5 is poor.

If the result is that I have no control over retract moves, isn't this a bug?......Would gtac be interested in getting to the bottom of this?


capnhook

Proud Member of the Reality-Based Community..

 

[cadcamnx]

hi

send me a part i go show how to it

hug !!!!

André

specialist and multi axis
development of IS & V
and postprocessors
member committee nxcam plm world
skype:andrenxcam7

 

[wmalan]

Ok I would create an additional extruded surface off that wall edge to use as a check surface just slightly concave to that wall. It can be just slightly set back enough that the tool gets a full cleanup of your drive surface but prevents the tool from wrapping around. In your operation you can set what the tool does when it encounters the check surface. I know it's not a click-your-done answer but sometimes geometry is screwy. Like Andre said if you could post that part file (or a small portion of extracted faces representing the retract problem) someone will nail it for you.

--
Bill

 

[capnhook]

Thanks. I will re-learn the many ways to jump through hoops, and skin cats.

It was nice to have the ultimate of predictable, stable, non-cutting moves in CATIA APT CAM. Hate to give that kind of control up.

Onward and upward......a bit sideways, too.

capnhook

Proud Member of the Reality-Based Community..

 

[markrief]

You keep asking about GTAC in this forum. Please contact GTAC directly to get your answers.
Thanks,


Mark Rief
Product Manager
Siemens PLM

 

[cadcamnx]

hi caphook


send me your file helps you stop it

Andre...

or report GTAC ;...

specialist and multi axis
development of IS & V
and postprocessors
member committee nxcam plm world
skype:andrenxcam7

 

[wmalan]

"Thanks. I will re-learn the many ways to jump through hoops, and skin cats.
It was nice to have the ultimate of predictable, stable, non-cutting moves in CATIA APT CAM. Hate to give that kind of control up.
Onward and upward......a bit sideways, too."

Like all things, NX does things different than Catia and vice versa. That said, I just completed a part that had some VERY finicky surfaces to drive. First done in NCL, then done in Catia, now done in NX. After a few trials, I found a way create a process to program them in ten minutes using NX. The Catia programmers had spent 2 days trying to get the proper output on each project. So it's a give and take.

Like Mark said, GTAC is your friend when you have problems. If you don't tell them, they can't help.

--
Bill

 

[capnhook]

Before I go off and contact gtac, has anyone else here noticed this type of erratic behavior that happens with non-cutting moves? It seems easily reproduced....

capn

Proud Member of the Reality-Based Community..