Equation curve trouble for cylindrical cam design

[cvs9QT7S]

Hi.

I am having some trouble getting the equation curve for a section of my cam to display the way I want.

Even though I have confirmed that my cycloidal function for the lift of the cam is correct (starting with a dwell section and ending in a dwell section) the curve just seems to follow a helix. 

I have attached som pictures to illustrate the problem.

PS. the reason I don't use design accelerator tor the cam is because of a problem where the roller path is not perpendicular to the center axis of the cam cylinder.

Here is the Equation curve in Inventor

This is the cycloidal function which I am trying to put in to the Inventor equation curve.

Any help is greatly appreciated 🙂

[karthur1]

I'm trying to understand what you are expecting the curve to look like.  Should it look like the "f" or "g" function from Geogebra?  Can you post the file you have so far?

[cvs9QT7S]

It should look like the g curve from geogebra, but I can see now that it is not pictured in the plot. However the only difference between the "f" and "g" function is the angular rotation of the rise/fall segment of the curve.

Here is a picture of complete cam curve, which I was planning to draw in four segments in a 3D sketch using equation curves:

I have also uploaded the Inventor file.

[karthur1]

I attached a Inv 2022 part below with the equation curves. When I am doing these equations, I find it easier to develop the equation in the parameters then copy/paste it into the equation curve.  Its easier for me to work out the units there. Also if I have trouble, I can evaluate just a portion of the equation and find the trouble.

Note: You are showing the value for function g at 135 and function f at 202.5 = 95.  The value I get at than angle is 93.345 deg.

Kirk

[JDMather]

---

@cvs9QT7S wrote:

PS. the reason I don't use design accelerator tor the cam is because of a problem where the roller path is not perpendicular to the center axis of the cam cylinder.

---

@cvs9QT7S 

Can you elaborate on this?

[cvs9QT7S]

There has already been another post about this, but I was not able to follow the solutions, if there ever was any. 

Here is a link to the post:

https://forums.autodesk.com/t5/inventor-forum/cylindrical-cam-generator-sides-of-cam-groove-not-alwa...

[cvs9QT7S]

Thank you for your reply.

Maybe I don't understand how equation curve works, but why doesn't the curves in your example start and end with zero slope in the interval determined by tmin/tmax? 
As you can see in the plotted functions from geogebra, the sine curves start and end at a constan y value. Why does this not translate into the equation curve in Inventor?

[karthur1 · Accepted solution]

My apologies. I made a units error in the sin function of the equations.  It should have been radians not degrees.

The values for the equations now are identical to the output from Geogebra. i.e. for function g @t=135, theta =95deg and function f@t=202.5, theta =95deg. 

When you get done with your project, it would be interesting to see  how you used these equations in your design.

Kirk

[cvs9QT7S]

Thank you very much for your solution it has been of great help to me. I have now made a complete 3D curve for the cam curve.

Now the next challenge is to make a cam groove that follows the path of the 3D curve. 

I have tried many different things but its quite difficult. I have also tried to make the equation curve sketch in 2D and projected the curve onto a cylinder with succes. However when i try to make a sweep afterwards the profile doesn't come out the way I want.

It seems that I run in to problems similar to the other thread I linked to, where the profile isn't shaped correctly.

I have attached the finished 3D curve as well as the part with the projected 2D curve.

[karthur1]

Have you tried to use the Sweep Solid tool along your 3D equation curve?

[cvs9QT7S]

Yes, it says that the sweeping path is not "tangent continous" but I can't seem to constrain the 3D sketch anymore than I have done, so I am not sure how to make the sweep work.

[karthur1]

When I try the solid sweep, I cannot get it to work either.  Tried several things, but no luck.  In the attached part, I am trying to solid sweep along just the portion of the equation curve and it still fails.  Not sure exactly why.

Inv 2023 file attached.

[karthur1 · Accepted solution]

I was finally able to get it to work.  I had to project the 3D curves onto the surface of the cylinder and use that for the path.  I also moved the start point to the 360° position.  It takes Inventor a few minutes to compute this curve.

For whatever reason Inventor does not like starting with the "3D Equation Curve2". I could not get it to work as one single sweep. Hope this helps.

Kirk

 

[cvs9QT7S]

I really appreciate the help you have provided so far. Unfortunately there is still a problem with the model as the "flats" of the cam groove are not perpendicular to the center axis of the cylinder which is actually the root of the problem.

This is what the design accelerator is also struggling with. I guess its not something that is easily done.

[karthur1]

---

@cvs9QT7S wrote:

.... Unfortunately there is still a problem with the model as the "flats" of the cam groove are not perpendicular to the center axis of the cylinder .......

---

Maybe I am not understanding what you are seeing.  From what I can tell, the flats are perpendicular to the center axis.

Can you show me where it is not perpendicular?

Thanks

[cvs9QT7S]

Yes I am sorry, I wasn't clear in my response. When I wrote "flats" I meant the surface which the cam bearing rides on, so essentially all the way around the curve.

[karthur1]

---

@cvs9QT7S wrote:

Yes I am sorry, I wasn't clear in my response. When I wrote "flats" I meant the surface which the cam bearing rides on, so essentially all the way around the curve.

---

Sorry, but I don't see where it is not parallel.  Attached is an iam with the cam roller following the cam profile.  Can you show me at what degree of rotation of the cam that profile is not parallel to the roller?

Cam and Roller Example

[cvs9QT7S]

It may be parallel, but if the surface is not perpendicular to the center axis of the cylinder (as it would be if you were to machine the profil in a cnc lathe for example) it is not going to work. This is a section view (in a plane through the center axis) of your part, and as you kan see the rolling surface is not perpendicular.

 

[karthur1]

---

@cvs9QT7S wrote:

It may be parallel, but if the surface is not perpendicular to the center axis of the cylinder (as it would be if you were to machine the profil in a cnc lathe for example) it is not going to work. This is a section view (in a plane through the center axis) of your part, and as you kan see the rolling surface is not perpendicular.

 

---

What you are seeing there is not what you think it is.  The reason those lines are not perpendicular with each other has to do with the geometry of the cut.  That section cut was made on the centerline at the 90deg position.  The line for that edge will be at an angle there because that is not the tangent point for the cam roller\cutter.

Below is a section at the same 90 Deg position.  Rather than being on the centerline, it is moved to the tangent position of the cam roller (or endmill).  From this, you can see that the edge is perpendicular to the centerline. 

Doing a solid sweep for this feature will result in the identical path an endmill will cut away.  It has no choice, but to be perpendicular to the centerline.

Hope this helps.

Kirk

[cvs9QT7S]

Thank for clearing that up for me, I had a suspicion that I wasn't looking at it the right way.