Helix......

[Sirius2]

Crimes, how do keep this one simple lol....

Right, I need to draw a spiral, a helix infact, much like a spring. I saw that Inventor has a 'coil' command, which I was counting on for doing this particular job (it doesnt get used much).

I start a sketch, draw a centerline, draw the coil diameter of 610mm, dimension the center of the circle to the centerline, exit sketch mode and choose 'coil'.

So far so good. The information I have at my disposal is the diameter of the coil centerline, and the hieght it needs to be, and that its one revolution, well, its actually slightly shorter, starting at an angle of 22 degrees (plan view) and finishing at 0 degrees. 0.911 revolutions I think.

However, I dont want 'flat ends' becuase it detrements the shape of the coil from helix by squashing it out of proportion at the bottom and top *in regards to the info I have* - YET, if I leave it alone with no flat ends, it sweeps the sketched circle up the helix - but the cross section of the coil isnt cylindrical (as the sketch is not perpendicular to the axis of the helix I defined in the coil dialog box).

I cannot use the flat ends option (which does *then* give me a true circular section all the way up) cos of the shape distortion and Im not astute enough to know how to construct a skeltal wireframe coil/helix and create a workplane without using the coil command.

I tried fooling the program by creating an extremely small 'flat' run off angle, but it doesnt compute it without being too large a flat/blend for the real world job. .

I tried it in MDT, and after reading the helix tutorial it actually creates a path and a workplane perpendicular to your information (hieght and revolution) where you *then* seperately proceed to sketch your 2d swept profile and sweep it- unlike inventor. The MDT plane always remains perp to the coil even when you edit it.

The trouble is that when you create views of helix's in MDT, you do NOT get a smooth representation of the coil, its rather hexagonalised (not viewres or facetres related) and therefore it looks utterly awful to the customer and even disjointed for the most part. Its unacceptable.

The Inventor tests Ive been doing come out better, but I cannot achieve the same thing as I drew in MDT, and exporting the MDT file into IV doesnt give me good results in Inventor - and besides, I need to be able to edit the helix incase of changes to the plans further down the project.

How do I draw this 'coil' in Inventor, given the hieght and revolution info, and maintain natural ends running off 'square' to the helix path and give a true cylindrical conformity all the way up the spiral, and not create an elliptical one as my initial sketch is not perpendicular to the helix angle?.

I cant create a workplane to what is currently anonymous information regarding the helix angle, as I only know hieght and revolution, so if I need to change the helix, the theory based initial angle will be wrong in context to the changes - and therefore again yield an irregular cross section. It needs to be planar/perp to the coil axis, terminating on the ht and rev info given.


Sorry if thats confusing, Its hard to explain. Hope you can clear this up for me!

Many thanks

Sirius2

 

[jlcochran1]

Interesting problem that if I understood correctly I was not able to find a solution to yet.

Alibre has a helix dialog box similar to Inventor, but adds an option to make the profile parallel or normal prior to the sweep.

SolidWorks takes an approach similar to what you described in MDT. First you use the helix command to create a centerline. Then you create a plane normal to that line. Draw a profile, then sweep the path. Cumbersome, but accurate.

My trial version of SolidEdge just expired so not sure how they handle.

In Inventor 10 I was not able to discover a way to keep the profile parallel to the sweep and maintain a parametric relationship for future editing. Hopefully someone with more Inventor experience will have a workable solution.

 

[Sirius2]

Thanks for that, I do find it hard to believe that a program as powerful as Inventor cannot create what is essentially a basic helix coil.

I appreciate you taking the time to actually try out what I was describing, becuase its only when you get doing it you fully realise where the problem is with Inventor.

Does anyone know how to do this? Sean, or some other uber Inventor user?

Maybe now you can use wireframe iges in v10, I can port a mdt helix to inventor and create a plane somehow, but its not parametric - which is the whole point of the program and quite crucial when you are working to bare-bone information to start the job ahead of time (needed to deliver on tight deadlines).


Myself and jlcochran1 cannot suss it - can you help master this inventor problem?


Sirius2

 

[MAGNUT45]

I work with rectangular, percision coil windings, maybe I could help.

What are the requirements? What is the highth? Could you eliminate the efforts so far and thin down the message to include just the details of what you need?

 

[Sirius2]

The data is subject to change you see, thats why it is suitable to a parametric modeller, hehehe.

These are architects drawings we have to work to, and lets just say they dont always show the truth or are done precisely OR have all the relevant information to turn their dreams into a reality through manufacturing drawings.

They have done a pretty good job, but as of yet I cannot be be confident of all the information.

What I essentially need to do, is draw a basic parametrically defined helix that works off the parameters of knowing the hight and revolution, radius/diameter of the centerline of the coil and the section shape based on a workplane that adjusts absolutely perpendicular to any changes to the helix coil 'pitch' angle.

Inventor doesnt seem to be able to do this. Or so I believe. The 'flat ends' option is not suitable as it distorts the helix when you only have hieght and revolution parameters to work with.

Any ideas how to do this and maintain a true circular section for the sweep at all times would be welcome! lol.


Sirius.

 

[Sirius2]

I have a cunning plan. I'll let you know how I get on. I think I know a work around. Its not ideal, but hopefully it will work.

I will be back.

Sirius.

 

[StuEllison]

hmm, I don't understand why your helix is not of unifrom circular section is your cirlce sketch in the same plane as your centreline?? can you not use flat ends and then make an extrusion based on a sketch perpendicular to your centre line?? have you tried bsaing your coil on two seperate sketches?? I'm not sure why you are unhappy with flat ends

 

[Sirius2]

I know its hard to describe, lol.

Yes the circular section is drawn at the same time to the centerline - inventor doesnt let you use a coil function without doing so. Thats the trouble.

The flat ends is not an option, becuase the only info I had was hieght and revolution. Using only hieght and revolution information in conjunction to using 'flat ends' doesnt maintain a true helix angle and therefore distorts and 'squishes' the true representation of a helic coil shape.

I feel Inventor is flawed in the proceedure for this function, and that it should have a seperate sketchplane placed perpendicular to whatever hieght and revolution data you have. Then, you should sketch on the plane and coil, this way nomatter what you change the parameters, the workplane/sketchplane always calculates itself perpendicular to whatever helix angle the ht and rev info gives.

What I ended up doing is using skeletal modelling, I think thats the term. I created an assembly, created a rectangular profiled coil to the parameters I had, then I made a new part, exited sketch mode and used a 3d sketch that 'included geometry' from the other part, and then created a workplane perpendicular to the 3d sketch.

Bit of a nightmare, but I got done what I needed to. I think they need to address this problem at Autodesk though.


Thanks for all your suggestions


Sirius2

 

[postmac]

Check out thread:

http://discussion.autodesk.com/thread.jspa?messageID=4374431