Multiple Tangental Constrained Radii

[kkeenn]

I'm dealing with a small sheet metal part that has multiple sequentially constrained radii (see attached picture, all radii are constrained). If I constrain the upper and lower surfaces with constrained radii I run into the possibility of the the geometry on the left occurring where one radii bows in and the other goes out. While I don't think its possible with a sheet metal part I would like to eliminate all doubt.

Is it legit to constrain one surface and then apply a profile tolerance on the cross section? Or are there any better suggestions?

 

[Belanger]

You could also add a note stating that the minimum thickness shall never go below X.XX. That way, all tolerances and notes would have to be achieved simultaneously.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

 

[BiPolarMoment]

It sounds to me that you don't even care about the individual radii rather than the thickness and resulting shape--why not dimension it the way that it matters to your needs? That would also better preserve design intent.

It's also probably a case of "that would never happen" but it's certainly preferable to dimension it such that it couldn't happen.

 

[kkeenn]

This is a simplified example. I actually have several (10?) Radii. I don't care where the tip ends but I do care about the radii are constrained and thickness

 

[drawoh]

kkeenn,

On a sheet metal drawing, you should show the material thickness, once. The inside and outside radii will not vary independently unless you use some other fabrication techhnique. Your thickness is controlled by the mill tolerances.

On complex shapes, you should use profile tolerances. Radii have a dimension, and a centre location. I find this hard to visualize with ± dimensions. With profile tolerances, you can make a tool that picks up the datum radius, and provide a comparison to the other radii, more in spirit with profile tolerances.

If you are making a tool to fixture to a datum radius, the opposite side of your clamp should be flexible enough to cope with expected variations in your material thickness.

JHG

 

[SteveMartin]

I would agree that the thickness should be dimensioned once, and only one of either the inner or outer radius in each pair - whichever one is more important to your part's function.