Three Face Blend

[cubalibre000]

Hi,
I'm trying to create a three face blend on the attached part as the image attached. following the PDF, but I'm not able to obtain the result as the image.
Can someone can give me the solution ?

Thank you...

Using NX 7.0.1.7 MP3 and TC8.1

 

[jarek30]

Hi,

I do not believe if it possible to do it this way. But I can be wrong...

 

[jarek30]

But I have acheived something. Look at the attachment. It's a move. I'm using NX 7.5.2.5mp03

 

[cubalibre000]

Sorry, but I'm using NX 7.0.1.7 MP3 and your option is not present.
This is new in NX7.5.

Thank you...

Using NX 7.0.1.7 MP3 and TC8.1

 

[mmauldin]

One way to get there:

 

[JohnRBaker]

Officially, NX did not support true '3-Face' blending until NX 7.5. Prior to that, there was a very limited tool which was part of the Face-Blend function, but it only worked in a few situations. Other than that, generally people had to depend on creating sweeps, such as has already been suggested, as a 'workaround'.

John R. Baker, P.E.
Product 'Evangelist'
Product Design Solutions
Siemens PLM Software Inc.
Industry Sector
Cypress, CA

http://www.siemens.com/plm

http://www.plmworld.org/museum/

To an Engineer, the glass is twice as big as it needs to be.

 

[cubalibre000]

Hi John,
so there is no solution using the face blend in NX7 ?
This shape is a cut in the plastic part, so it's very difficulty use the trim body or similar workaround.

Thank you...

Using NX 7.0.1.7 MP3 and TC8.1

 

[JohnRBaker]

There is NO practical '3-face' blend in NX 7.0.

John R. Baker, P.E.
Product 'Evangelist'
Product Design Solutions
Siemens PLM Software Inc.
Industry Sector
Cypress, CA

http://www.siemens.com/plm

http://www.plmworld.org/museum/

To an Engineer, the glass is twice as big as it needs to be.

 

[FrankSwinks]

Attached is a NX7 part with both an approximate solution and an exact solution using two rolling ball face blends. The dialog is shown in image1.jpg. This method requires a tangent edge. My approximate solution is shown in image2.jpg. The approximate solution part is also included in the zip file. Now what makes this solution approximate is that the tangent line (in blue in image3.jpg) needs to be equidistant from the two blend edges for the final blends to be true 3 face blends. For the approximate case I have used the centreline (which is easier to create). If you are able to measure the two red lines you will see they are not equal in length. It is not a difficult journal/program to generate the correct tangent line. For this case with planar faces it is not difficult to create the correct tangent edge. The part with this accurate solution is also included.

Hope this is of use.

Frank Swinkels

 

[cubalibre000]

Thank you FrankSwinks for your alternate solution.

Thank you...

Using NX 7.0.1.7 MP3 and TC8.1

 

[mmauldin]

I took a look at Frank's models in 7.5 and added a true 3 face blend along with my original swept feature version. Attached are jpegs showing the results. None of the models are particularly close to the true 3 face blend. I then went back to NX6 and created the blend using sketches that take into account that the side walls are not perpendicular to the upper face. Using this method the to determine the line of tangency results in the jpeg file "mike2.jpg" which appears to match a 3 face blend very closely.

Sorry about the conversion to inch units. I imported a parasolid rather than opening it directly.

 

[FrankSwinks]

OK. I assumed (incorrectly it appears) that the angles beween the (to be) trimmed faces and the (to be) deleted face were equal hence the equal lengths used for the locating the tangent line. When the included angles for the rolling ball are different for the two blends, further calculations are needed to determine the correct tangent line.

Frank Swinkels

 

[FrankSwinks]

I have thought a little more about this and now realize that my solution does not create a true 3 face blend. I appologize for my error.

Frank Swinkels

 

[FrankSwinks]

To convince myself that it is possible to do it with the method I previously suggested I had a further go since the two faces are at different angles it took a little more work to achieve the correct tangent edge. Once achieved I believe it is the correct outcome.

Frank Swinkels