Defining the orientation of a helix curve -NX5

[RobLN]

I could do with some advice on the construction of the helix curve. If you go insert>curve>helix then OK the helix is defined on the 0,0,0 of the coordinate system.

I want to start the helix on an angled plane relative to the datum coordinate system. However, seems impossible to define the orientation within the command. If you go 'specify the orientation' a useless dialog box pops up. There is Name = (with no explination of what is supposed to be entered) and the system is asking to select a Z axis but there is nothing selectable in the model window. (even though the selection filter is set to 'no selection filter')
As this method failed I also tried to define a plane on a helix curve (% along the helix to control it). Then sketch a circle on the plane, with centre constrained to the helix curve. Then sweep the circle the helix to the required shape, again this failed too. See PDF file for NX5's wierd interpretation of what should be a fairly basic operation..

Any advice greatly appreciated.

 

[braddles90]

I'm at home at the moment so I can't try this now but im 99% sure this would work. Can you just orient the WCS to the co-ordinates you want to (say (3,4,5) and then orient the z-axis to the direction you want, then create the helix?
Your datum plane should be relative to the absolute datum, not the WCS so you should be able to move it around without affecting everything else, then once youve created the helix you want just reset the coordinate system back to absolute.

 

[phillpd]

As far as I'm aware, the positioning of a Helix (Insert->Curve) is not parametric.

I would suggest you use a Law Curve as you have far greater control.

Specialty Engineered Automation (SEA)

http://www.sea4ug.com

a UGS Foundation Partner

 

[cowski]

The above posts are correct. The helix command is a strange mix of associative and unassociative parts. As braddles90 said, you should orient your WCS where you want the helix before you start the command.

See this thread for some workarounds that may or may not be helpful to your situation: thread561-164528

 

[RobLN]

Thanks for all your replies. I will try out the law curve method but I have tried that before and, bearing in mind I've had no formal training on that, it seems quite complicated to get a Helix shape with the law curve. If anyone's got a step by step on how to make the law curve helix feature in the part navigator that would be great.

In the end I tried a work around for my spring (an extension spring with two hooked ends)where I left the helix static and made the perpendicular sketch plane with the hooks on it rotate around the helix to adjust - worked OK. I also used bridge curve to link between the hooks and the coiled portion of the spring. Is there another more 'engineered' way to create this rather than adjusting the bridge curve so it 'looks right'?

 

[hudson888]

If your spring section is circular as it is for the majority of cases use a tube rather that a sweep. Sweeps using long an torturous paths are naturally more complex and therefore difficult to control. The tube function will give you a solid if you specify a zero ID value, with a positive OD value.

A helix is nothing but a canned example of a law curve the axes you need to control in most cases are easily enough manually oriented until you have to change their orientation. The truth is that in defining springs as components we rarely of ever need to reorient any of those elements so it simply doesn't come up.

Best Regards

Hudson

www.jamb.com.au

Nil Desperandum illegitimi non carborundum