surface to solid problem

[michaelmould]

Hi,

This has been troubling me for a couple of days now. I'm fairly new to surfacing, so I apologise if this is an obvious mistake.

I have modelled a surface which I want to turn into a solid using the thicken tool, but I can't get it to work. As far as I can tell the surface is completely tangent continuous.

I have also tried using the offset surface tool, again with no luck.

Any advice you can offer would be greatly appreciated.

Thanks in advance!

Mike

(NX6)

 

[KevinDeSmet]

Trouble like this is almost always caused by the minimum radius of curvature, that's the smallest circle that can fit between two pieces of your surface.

Unfortunately this is always tricky to deal with.

I have Solidworks here at the office and I opened up your file, now I could be getting import/export issues but I ran a check and it found a minimum radius of curvature of 0.02mm - now there's your problem

It's at the area where the "ruled" surface is hanging off the big swoopy surface. The big surface is curving grotesquely much near the bottom and it shouldn't do that, that makes thickening it impossible.

Hope someone else can help out, more specific to NX.

CSWP-Surf

 

[ingallspw]

If the surface curvature is too complex, folds over itself, or changes vectors, the thicken will almost always fail.

Keegan

 

[ingallspw]

I opened your model, I too tried to make a surface fold as you have. You will need to either create a whole new bottom surface or get rid of the fold and work from there.

Keegan

 

[hudson888]

Okay well Kevin is definitely in the right ball park. Have a look at the attached file and you'll see I have taken a completely different approach suggesting a method that might deliver something of the result you sought.

You forgot to say how thick you wanted it so I just guessed 3mm might be okay. You might want to replace the simple radius edge blend with a studio blend or conic, but then as Kevin suggested you'd need to use analysis tools to ensure you maintained a minimum radius.

You my shape isn't the same as yours, but I could push it anywhere you wanted to go again given time and perhaps a little latitude.

The main reasons your surface doesn't work are that the sweep around the outside was by no means tangent continuous at all, and your central surface was built using what you probably didn't realise was impossible technique. Surfaces operate as a grid of sections that cross one another so the cardinal sin that you made was to attempt to define a shape using curves that meet with tangency. Whenever the primary and cross curve form either a point or are continuous there is no effective control over what the shape does. Hence areas within your sheet wind up having very sharp direction changes that were never going to thicken.

Let me know if you like what I've done and whether it adds to your understanding. I know that this answer is perhaps quite different to what you expected so I wouldn't be surprised if you had some misgivings about it.

Best Regards

Hudson

www.jamb.com.au

Nil Desperandum illegitimi non carborundum

 

[hudson888]

Well that didn't attach did it!

Take 2. File Attachment.

Best Regards

Hudson

www.jamb.com.au

Nil Desperandum illegitimi non carborundum

 

[michaelmould]

Hi guys,

Thanks for your input. Hudson, thanks for your model. It explains a lot, and having looked at my model once agin, I can see where the surface is pinching where the tangential primary and cross curves meet.

As I mentioned I am quite new to surfacing, so am open to any suggestions from anyone with more experience than myself!

I'm going to have a go at remodelling the chair taking your advice, and maybe build some construction surfaces to produce the rail so I have a more control over the blend and the depth of the rail as I want it to vary at different points along the edge.

This started as a quick model to go into a room set for a rendering I am doing, but its amazing how easily one gets distracted when you want to do something correctly!

Thanks again for your help.

Mike

 

[SiW979]

Hi Mike

I've attached another variation, whilst the majority is based on the same method as Hudson's, I have used a totally different technique to create the lip around the edge by extracting a face (at time stamp) offsetting the edge curve in face, trimming the inside of the extracted sheet away, then thickening the lip, uniting it and adding seperate blends.

See what you think



Best regards

Simon NX4.0.4.2 MP10 - TCEng 9.1.3.6.c - (NX6.0.3.6 MP2 native)

http://www.jcb.com

Life shouldn't be measured by the number of breaths you take, but by the number of times when it's taken away...

 

[SiW979]

Mike

Ignore the last model, I forgot to delete a lot of the other stuff away which may confuse you and I had also made a little mistake, this one is right now.

Cheers

Best regards

Simon NX4.0.4.2 MP10 - TCEng 9.1.3.6.c - (NX6.0.3.6 MP2 native)

http://www.jcb.com

Life shouldn't be measured by the number of breaths you take, but by the number of times when it's taken away...

 

[hudson888]

Just in terms of construction if nothing else why not just thicken to 13mm add on the 8mm blend then shell the body from the rear to 5mm and blend sharp edges 2mm to finish. Same result just as simpler method.

My example was admittedly headed in a different direction and not entirely the most straightforward thing itself so I'm uploading a nicer version where I have added some extra shape control courtesy of added cross curves and a little massaging of the base curve's shape. To finish some styled blends will really pick out the edges nicely and make a better job of things.

Now if you wanted to ask why I messed about with larger flanges that were extruded then that's because Panton's original chairs had a larger flange with a shaped edge and bigger edge blends that anything we're doing here. I think that by adding an extra sketch or two and taking some time over the blends one might see how it could be quite easy to get reasonably close to his original.

Best Regards

Hudson

www.jamb.com.au

Nil Desperandum illegitimi non carborundum

 

[michaelmould]

Hi Simon, Hudson,

Thank you so much for your posts, these are most helpful! I can see I have a lot to learn not only about practical surfacing, but also about theory behind it too. I now appreciate why the quality of the curves is so important in creating quality surfaces.

Thanks once again, I will post my efforts once I have more to show. 45 minutes at lunchtime just isn't enough to get things done!

Cheers,

Mike