Datum structure for thin sheet metal part

[310toumad]

I have a part that is .036" thick, which essentially looks like a ring with tabs around the circumference of the O.D. (part is about 2.3" diameter on O.D., 1.75" I.D.). The tabs are angled slightly to give a lead-in because the part fits inside a tube with a slight interference. The flat ring sits against the surface of another part inside the tube. There is also a cylindrical part that passes through the I.D. of the ring, that fits inside the tube as well.

For setting up the datum structure, in my mind datum A would be the flat surface of the ring that sits against the other part in the tube. Datum B would be the O.D. of the tabs since they determine where the center of the part is when installed. Then I can establish a positional tolerance for the I.D. relative to these datums to ensure there isn't interference with the cylindrical part that passes thru the center of the ring. The issue I am having is thinking of a way to relate datum B back to A. The only thing that would makes sense is to define an orientation of perpendicularity, however the part has no depth to it so that seems non-sensical to me and would be difficult to even measure.

Another thing I thought about was eliminating calling out the flat surface as datum A, and just using the O.D. of the tabs as the primary datum with no secondary. Does this make sense since the part has no depth to it? Really the concern is establishing a concentric relationship between the O.D. of the tabs and I.D. When the part gets installed in the tube, and mates against the surface of the other part, it will get "flattened" out one way or the other, so perhaps that flat surface of the ring is not critical in terms of being defined as a datum.

 

[chez311]

Any chance you've got a drawing/photo/sketch to share? Also what standard are you working to - ASME?

 

[310toumad]

Here is a photo of a similar part.

 

[AndrewTT]

Do you want the part inspected in the free state, the restrained state, or a combination of both?

 

[310toumad]

Most likely in the free state, the only restraining forces would be the interference between the I.D. of the tube and the O.D. of the tabs on the ring, which isn't enough to deform it enough to affect its function.

 

[chez311]

When the part gets installed in the tube, and mates against the surface of the other part, it will get "flattened" out one way or the other

Sounds to me like it does deform a notable amount during assembly.

 

[310toumad]

What I meant by that was, if the ring is not "perfectly" flat to begin with in its 'whatever has been produced' state, I'd expect any variation there to be insignificant, such that when its being pressed against the flat surface it would only bend a relatively small amount.

Which leads me back to my original question. Is defining that flat surface an unnecessary level of control on this part? Would the primary datum axis as defined by the O.D. of the tabs be sufficient?

 

[pylfrm]

Is defining that flat surface an unnecessary level of control on this part? Would the primary datum axis as defined by the O.D. of the tabs be sufficient?

The large flat surface that contacts the mating part should almost certainly be the primary datum feature. This matches the functionality you describe. If the tips of the tabs are all in approximately the same plane, they will not be able to constrain the necessary rotational degrees of freedom to serve as a useful primary datum feature.

Even if it does not match the intended inspection method, you may want to consider specifying the secondary datum feature to be the tips of the tabs as restrained within a cylinder of a certain diameter. Again, this matches the functionality you describe. It should be easy enough to demonstrate a suitable correlation between restrained and unrestrained measurements, allowing most of the inspection to be done unrestrained if desired.

How do you intend to control the tips of the tabs to ensure the proper interference and engagement with the inside of the tube?


pylfrm