runout vs concentricity question

[tactical14]

If I mount the following object on a V block at point A and put a indicator at point B, what am I measuring? Basically I want to see if the two diameter's center lines are collinear.

 

[TheTick]

Sorry. I asked you to use imagination. That might have been a stretch. An actual corkscrew would be extreme, but a very subtle one is quite possible (and even common in some applications).

Another example:
Posted here is an example of opposing ellipses which would show zero runout when measured on a V-block. Again, this example is exaggerated for effect.

It is also possible to have shapes that appear to have uniform diameter when measured with calipers at different orientations but are not round.

 

[pmarc]

TheTick,
It would be good if your attachments vere in formats readable to all of us here. I do not have Solid Works, so can just imagine what your "another example" is about.

You seem to stick to V-block issue. The thing with it is that if you are sure that datum feature experiences significant form and/or size errors (within stated limits of size if we are in ASME world), because you previously checked it, v-block should not be used for datum axis simulation, exactly due to the reasons you mentioned. In reality it happens quite often that datum features are of sufficient form and size quality to enable v-block usage. Sometimes v-block(s) is(are) the only reasonable choice.

I have a feeling that v-block was not the clue of OP's question. Of course it is very good that the caution was urged by you and others, but did it bring him any closer to understand what can be measured by dial indicator on feature B? I do not think so.

 

[TheTick]

I seem to have read the original question and not gone off into the weeds.

I've been over the concentricity vs. runout thing many times in my career. There are thems that gets it and thems that refuses to. Not much in between.

Sorry about the format. I tried to post images but it didn't take. Posting composite image of the model.

Honesty may be the best policy, but insanity is a better defense.

http://www.EsoxRepublic.com

-SolidWorks API VB programming help

 

[pmarc]

Technically, it does not have to be true that for these ellipses zero runout error will be inspected.
If the part stands still and the indicator rotates around datum axis, the error will be there, don't you think?

 

[CheckerHater]

The part will show HUGE runout but zero concentricity
(Sorry, I know I promised to keep quiet)

 

[TheTick]

Part was designed to demonstrate ZERO runout when checked in 90° V-block. This is why V-blocks are bad, m'kay?

Are you going to rotate the indicator around the axis? How are you going to set that up without a mandrel? If you had a mandrel, why are you using a V-block?