Good design or not

[[SH...]]

I checked the perpendicularity( yellow colour), cylinder as datum and plane as feature I got 0.153, but I check it plane as datum l got only 0.005. My doubt is that it is a good design asking perpendicularity w.r.t cylinder. In my knowledge it should be w.r.t plane.

IMG20200519161057.jpg

[[Cl...]]

The cylinder isn't deep enough to be used as a Datum (depth 3x diameter)
I think that's primarily why you are seeing such a dramatic difference between
both methods.

[[SH...]]

Please sign in to view this quote.

This is a standard??

[[Cl...]]

No, just a general knowledge rule of thumb.
Something I learned here on the forum.

[[Me...]]

I've always pointed to Section 4.8 to properly explain this issue when an Engineer puts illogical callouts such as this one on a drawing.

I see this problem be most common with perpendicularity of a face to a bore.

[[Ri...]]

You should see if they will allow you to make a composite datum of P-Q. That will make the datum axis a lot more robust. The way that perpendicularity callout is drawn is essentially an Axial Runout.

You can show them how unstable the current drawing is by showing the Perpendicularity the way they have it drawn, and then making a subsequent one where the datum and feature is swapped.

As for the 3x length to diameter rule, it is a great suggestion, but it's hard to follow when a majority of parts don't allow for this. The only thing you can do is explain the error to the design engineer, and hope he changes datums around, or increases the tolerance.

[[Ow...]]

Illustration attached.

[[Da...]]

If it function that way then it is right, but more then likely it will conform to the plane and twist the datum in the bore. The large surface area of the plane will become primary in the assemble.

I had a callout that was a 1 inch deep bore with a 39 inch diameter plane surface that was .0005 perp to Bore (Datum -E-)

That is 39X calculation so for ever .0001 error the bore had calculates to perp of .0039.( The bore would need to be with in .00001282 and plane perfect to be in spec.)
But during assemble the plane is going to be come Primary because it's 34 bolts that hold it to the cover will over power the bearing. The bearing in that bore is going to have to take the error in misalignment.
It took time to explain to the engineering team but it was changed in the end.
Good Luck with getting something changed.
If it is DOD or Aerospace its not happening more then likely.

[[SH...]]

It is an automobile part, eventhough my perpendicularity get wired result, my customer is accepting the part. I think they also get same result...

[[Ri...]]

I had a white paper from a seminar a decade ago? Not sure what happened to it.
The white paper was for CMM programming and the title was: Short Unstable Datum's or Features.
It explained in great depth (Math beyond my capability) and enough laymen's lingo to follow along and understand it.

In a nutshell, the brief description Clarke Gilbert posted is true.
Typically when you run across this scenario simply swapping which feature is the Datum will render the correct answer when compared to a layout value.

[[Am...]]

Please sign in to view this quote.

If you ever run into that paper id be interested to see it.

[[Ma...]]

This is one of the most common GD&T problems I see. How would everyone deal with this?

Report results as on the drawing?
Report results using perpendicularity of the cylinder with the plane as the datum?
Report both?
Don't measure the part until the drawing is changed?
Something else altogether?

I have probably done all of these at one time or another over the years.