Equation curve problems

[leonardo.k]

Hey guys, I just performed some tests with equation curve, and I think something is not right and I can't see where:

I made 2 functions

R=100mm

m2=0,25ul

This is the curves:

(R+m2*R)*cos(m2*t)-m2*R*cos(t+m2*t)
(R+m2*R)*sin(m2*t)-m2*R*sin(t+m2*t)
tmin=0
tmax=360

(R+m2*R)*cos(m2*t)-m2*R*cos(t+m2*t)
(R+m2*R)*sin(m2*t)-m2*R*sin(t+m2*t)
tmin=2*360
tmax=3*360

(R-m2*R)*cos(m2*t)+m2*R*cos(t-m2*t)
(R-m2*R)*sin(m2*t)-m2*R*sin(t-m2*t)
tmin=3*360
tmax=4*360

(R-m2*R)*cos(m2*t)+m2*R*cos(t-m2*t)
(R-m2*R)*sin(m2*t)-m2*R*sin(t-m2*t)
tmin=360
tmax=2*360

And this is the result:

I can't extrude this profile, even in surface mode, but if I use this same profile with a patch command and use Thicken/offset I can create a solid form:

I'm using Inventor 2022.

I'm pretty sure that is a problem with this equation, but I can't figure out where is the mistake.

Thanks a lot!

PS.: Already look at some post in this forums, tried to apply some suggestions but the result is the same.

[Christopher2222]

Why are you using multiples of 360?  tmin 2*360 tmax 3*360 would be a full revolution through 0 degrees at the second revolution to 360 degrees at the start of the third revolution.  I believe this is where your problem is. 

[WHolzwarth]

Two comments:

- Equation set 1 is the same as set 2, set 3 is the same as set 4.

- Perhaps you only need to set coincident constraints between the curves.

Best way is uploading your file

[Christopher2222]

The equations are fine, they work out in math software.  It's almost like in autoCAD if the lines are separate you can't extrude it.  Almost like you need to join the lines. 

I suggest perhaps making a simpler equation curve and see if you can get that to work, see if it's an inventor issue or a setup issue. 

Possible it doesn't like the variable substitution.  Replace m2 and R with their values when setting up the equations and see if it works that way without the m2 and R variables.

[johnsonshiue]

Hi! I suspect this has something to do with small gap or high curvature at the coincident points. I am sure the curves are mathematically correct. The issue here is that Inventor has to generate the geometry within 0.00001mm tolerance (not zero tolerance).

Boundary Patch tolerates such profile better because it approximates the boundary curves within looser tolerance.

The workaround is to use BP to create the surface. Then thicken it to solid. Or, project the BP edges to another sketch.

Many thanks!

[leonardo.k]

Hey guys, thanks for the help.

I tried all the suggestions and still have the same problem.

If I use BP and then Thicken/Offset, I can't even create a sketch on the solid plane, I try to perform a Sweep in the edge, error too.

This is a weird curves, I used sketch doctor to try to find some problems, and nothing, everything is fine on sketch, all points are constrained with others.
Well, I will try to find something to fix this issue.

[Christopher2222]

Inventor 2015 can't even get by your first two equations. 

I get the error "The curve is irregular at parameter value 0.000000"

and your 2nd set of equations spring this error on me.

    "The curve is irregular between parameter values (720.000000, 1080.000000)"

I thought it might have something to do with the boundaries or tolerances as Johnsonshiue suggests.  Actually the reason is here https://forums.autodesk.com/t5/inventor-forum/equation-curve-irregular-at-parameter-value-0/td-p/548... and was fixed in 2017.  Sounds like the same type of issues are still occurring in 2022.  So boundary edges aren't defined and therefore doesn't create a closed object that can be extruded.

[Christopher2222 · Accepted solution]

Ok I've solved it here

So it exactly has to do with the tolerances.  The workaround is to add 0.1 to tmin and subtract 0.1 to tmax on each pair of equations and your equations are fine, so ...

set 1 - tmin=0.1  tmax=359.9

set2 - tmin=720.1  tmax=1079.9

set3 - tmin=1080.1  tmax=1439.9

set4 - tmin=360.1  tmax=719.9

then use the coincidence constraints on all the end points of the 4 curves to connect the lines together.  After that extrude will work.

This type of solution will work for all issues regarding EDC (equation driven curves) that have errors mentioning irregular curves at, or between, specific points.

I initially tried using line to join the points but inventor wouldn't make a line that small.  It appeared that it did but when I zoomed in later I discovered some lines weren't created. 

[WHolzwarth]

Hmm. Some kind of mystery for me.

I've placed the formulas in an Excel File and got curve points, as can be seen in the diagram.

 Can I have a look at the IPT file?

[Christopher2222]

I can send when I have access.  Not sure why you are using Excel???

Using the equation editor for parametric equations and entering the values the OP had and using my limits you should end up with 4 curves in inventor.  The ends of each curve are not quite touching, and of course, extrude won't work on areas that aren't closed.  

Use the equation curve button

(for 2015 it's hidden under 2dsketch->line submenu) you will get your equation editor where you can enter the equations - no need for excel

[WHolzwarth]

I know about using equation curves. That's not the matter.

I simply wanna see the IPT, and check the curve equations with the initial formulas on top of this thread.

[Christopher2222 · Accepted solution]

Oh, ok. Won't be able to add till the end of the day.
I only substituted the values and changed the range then applied coincidence constraints.
125*cos(0.25*t)-25*cos(t+0.25*t)
125*sin(0.25*t)-25*sin(t+0.25*t)
tmin=0.1
tmax=359.9

125*cos(0.25*t)-25*cos(t+0.25*t)
125*sin(0.25*t)-25*sin(t+0.25*t)
tmin=720.1
tmax=1079.9

75*cos(0.25*t)+25*cos(t-0.25*t)
75*sin(0.25*t)-25*sin(t-0.25*t)
tmin=1080.1
tmax=1439.9

75*cos(0.25*t)+25*cos(t-0.25*t)
75*sin(0.25*t)-25*sin(t-0.25*t)
tmin=360.1
tmax=719.9

[leonardo.k]

Hey guys,

The IPT file, Inventor 2022.

[leonardo.k]

This solve "part" of the problem, still not able to use edges to chamfer, fillet or sweep, but this help a lot!

Thanks

[WHolzwarth]

Ok. Now I see.

😁 What a difference coordinate system indication makes ..