Measuring runout on a shaft relative to two datum surfaces... part 2

[supergee]

Hello All,

Following thread1103-507627 that has been closed since, I had additional questions on the evaluation of the run out when two datum with two diameters are called in the drawing. I understand the limitation of using V-blocks to setup the inspection. This being said, when calculating the height difference between both v-block, the diameter used in the 1.414 x (D-d)/2 should be the actual diameter or the basic diameter?

I would be inclined to says the actual diameter but then if the datum represents the condition where the part is used, the mating elements would normally guide the part and not the other way around... I don't know if that makes sense. Plus, on would have to reposition the v-block for each part which would make a very tedious inspection.

Thank you all.

 

[Burunduk]

I would also say the actual diameters.
The diametric (in the end - radial) difference would affect the adjustment movement distance between the V-Blocks, and you wouldn't want the axis of rotation in your arrangement ending up skewed relative to the runout measurement direction because of the difference of the actual diameters from the nominal ones.

 

[3DDave]

supergee,

If the v-blocks are not of a height to make the mutual axis parallel to the inspection table and one is measuring total runout with a dial indicator then any error in the difference in heights will be seen as an error in total runout in the axial direction. It won't affect runout measured in the circular direction.

Eaxmple: If the part is 20 inches between the datum features and the height difference from perfect is 0.015 and the width of the toleranced feature is 2 inches then the error in the axial direction will be (0.015 *2/20) inches or 0.0015 If this difference is from the difference from nominal size or median size of the shaft datum features that represents a a total diameter divergence of 0.030 inches (I think you should draw up a diagram yourself or use CAD to create a model to change the diameters to check your specific case. For what I am paid here, I'd rather you do the checking to be sure) Obviously if the mutual variation of the datum features from nominal is, perhaps. 0.001 then the contribution could be on the order of 0.0001 inches to total runout.

The other problem with being the wrong height is that the shaft's datum features will sit on one edge or the other of the v-block or the end of the shaft rather than making good contact with the v-block as if the datum feature is a narrow ring. You can measure the gap under the lower end to know how big a shim to use to lift the low-side v-block.

I think I recommended collets in the previous thread. They are automatically adjusted by installing them. They represent the more typical fit rather than a v-block. The goal is to eliminate the errors in the measuring scheme and see only the variations of the part.

 

[supergee]

Burunduk, 3DDave,

Thank you for your answers! It confirms my original theory.

I am trying to mix GD&T Design theory and Metrology in the same course in order for future designers to understand how the GD&T work and will be inspected and make the subject more "real".

As I said, I understand the collets to be the better option but our college has budget restrictions and we have to choose between updating our 1990's machinery or our 1980's metrology equipment. Lets just say that we will still be using vernier scaled calipers for the time being: none of those fancy schmancy electronic displays for OUR students.

I wanted to show the students what to do if the're in a company with low budget... like our college for instance.

Gee