Conical vs. cylindrical fillet

[bnothstine]

I am trying to put an R.015 conical fillet TYP 4x. The other rads I put on the sketch. When extruding, they become conical. When I tried that with the R.015, I get an error: LOP_TWK_NO_VERT: no solution for a vertex.

If I fillet the part, I get cylindrical unless I do a variable rad, but how do I make it taper out at approx. .122" below the top surface? If I put in start at .000 and end at .015 it goes all the way down the part.

I can program my CNC grinder to make a cylindrical rad on this carbide insert, but I would prefer a conical one, and I run into this problem fairly often.

Thanks in advance for your help!

 

[HaroldL]

Is this what you're attempting to create?


Looking at your part, it appears the sketch based radii blend to zero after the Draft is applied. To get your feature based radii to "taper out at approx. .122" below the top surface" you may need to make a break in the edge. A miniscule cut would work.

Another method may be to extrude your part to the length needed for the variable radius, apply the radius, then create another extrude from the face to the final size needed. This is the method I used to create the part in the image above.

With Draft applied it can become a bit tricky to match the draft faces. Using the Draft function on the second extrude does not seem to work, but you can use an extrude cut to match the faces.

 

[bnothstine]

Thank you.

"Another method may be to extrude your part to the length needed for the variable radius, apply the radius, then create another extrude from the face to the final size needed. This is the method I used to create the part in the image above." does work.

However, sometimes the calculations to achieve the "taper out at approx. .122" are exceedingly difficult. The R.03125 will taper out at: .03125/tan(11)=.1607 etc. so it is not exact to do the double extrusion. And it does not generate the compound angle (arctan (tan11/sin45)) underneath the rad. See ANSI standard carbide insert attached.

Any other ideas? I think the solution lies in the variable rad formula but I have not been able to get it to work correctly.

 

[HaroldL]

...
However, sometimes the calculations to achieve the "taper out at approx. .122" are exceedingly difficult. The R.03125 will taper out at: .03125/tan(11)=.1607 etc. so it is not exact to do the double extrusion.

Okay, now I'm confused, do you want it to be approximately .122 or Exactly? (what's the tolerance?)

And it does not generate the compound angle (arctan (tan11/sin45)) underneath the rad. See ANSI standard carbide insert attached...

Where are you measuring the angle? Can you Redline the model so I can understand the "underneath" comment?

 

[bnothstine]

Okay, now I'm confused, do you want it to be approximately .122 or Exactly? (what's the tolerance?)

Thanks for your patience. I really appreciate the help. Hopefully the attached will clear it up.

Really the approx vs. exact was just a "for example". In this case the .03125 radius tapers to zero size at .03125/tan11=.16077 from the top of the insert.

The compound angle should be (arctan (tan11/sin45))=15.57deg but measures 15.37 on the part.

I know it sound insignificant, but I work in tolerances of microns and 1/1000th of a degree. It has to be perfect.

 

[HaroldL]

... The compound angle should be (arctan (tan11/sin45))=15.57deg but measures 15.37 on the part. ...

:? :? When I enter the formula in my Casio FX-260 calculator I get 15.37036237 as the result.
The keystrokes I use are: (/ = divide)
11 TAN displays .194380309
/ 45 SIN displays .707106781
pressing the = equal key displays .274895269
pressing the Shift and Tan key (Tan-1) displays 15.37067237

--
edit...In order to get 15.57 on the model I need to set the draft angle on extrude to 11.146145 or rounded to 11.15.

 

[bnothstine]

whoops. must have fat-fingered a key, or misused the measure tool.

In this simplistic example, yes it corresponds, but in the original file I posted, I can't make it work. The fact remains. I can't make the model match what the machine produces.

 

[Ralf]

Brian,

The fact remains. I can't make the model match what the machine produces

You can create this model, and I would recommend to use the 3D features that Alibre / 3D CAD software offers,
combined with the experience of handling these "tools/features", like:
Planes and axes at an angle -> Sketch - Sweep Cut feature...

 

[bnothstine]

Of course you are correct. I should be able to make these features.

That is really why I posted my problem here, hoping for guidance to find a solution. I did not mean to imply Alibre does not have the capability.