Is this a method of GD&T simultaneous referencing?

[ronj100]

Not sure how to interpret this. (see what is in green oval)
At first I thought it was wrong.
That possibly datum -c- should be by itself with its own FCF using just datums A & B, and then the other three holes with a feature control frame referencing back to A, B & C.

But then someone told me that was legal because it was using simultaneous reference positioning.
However, after reading up on simultaneous referencing, I'm not sure.

See attached print

[Kelly_Bramble]

ASME Y14.5-2009, paragraph 4.9 "Simultaneous Requirements" ...

No, this does not appear to be a valid definition. I'm sure that there may be some folks whom might argue a gray interpretation where all of the hole features are Datum "C" and then measured to each other but - don't. You'll likely confuse the issue.

One of the holes should be exclusively specified as Datum "C" and the other three oriented and located back to Datum's A, B and C.

[Belanger]

If all holes are the same size and of the same importance, I wouldn't even use just one hole as the datum feature. Just make the 4-hole pattern create datum C. This can be done by dumping the reference to C(M) in the green oval, and then moving the datum feature symbol to hang under that feature control frame. (Then, you'd probably want to and the MMB modifier to datum reference C in the all-around profile callout.)

[Kelly_Bramble]

I knew you was going to say that...

So, let me corner you. Per. your recommendation, how does one establish datum "C" from the four holes using a CMM?

And... what are the chances that the quality metrologist will get confused with the specification? and then just do what they want..

If all holes are the same size and of the same importance, I wouldn't even use just one hole as the datum feature. Just make the 4-hole pattern create datum C. This can be done by dumping the reference to C(M) in the green oval, and then moving the datum feature symbol to hang under that feature control frame. (Then, you'd probably want to and the MMB modifier to datum reference C in the all-around profile callout.)

[rpolleys]

CMM will need to construct a hole pattern as a set an the set can have a hole pattern as a datum , the tricky part comes in when the software needs to act in this case as a gage simulator and software I use does this.
As far as clarity for interpretation add note to drawing "C" Datum is 4 ID pattern

[Kelly_Bramble]

Let's examine the two specifications..

Specifically, the CMM will establish Datum A by contacting many areas/points (minimum of three) and deriving a datum "A" plane. Then we establish Datum B by contacting a minimum of six areas (points) on Datum feature 'B". Three at approximately 120 degrees apart toward the top of the cylinder and three at approximately 120 degrees apart towards the bottom of the cylinder. This will establish an axis and with that axis we get two datum planes - which are not constrained and can rotate about datum axis “B”.

At this point we have three axis and three datum planes where two of the datum planes can rotate about datum axis "B".

Datum "C" will now be used to orient the two datum planes that can rotate about datum feature "B" and therefore lock down the entire six degrees of freedom for our coordinate system or DRF.

If only one hole is used as datum "C" – we would then establish three areas of contact 120 degrees apart toward the top of the hole and then three areas of contact 120 degrees apart toward the bottom of the hole to establish an axis which the existing datum’s planes rotating about datum B will be oriented and locked down to.

The location of Datum C can be simultaneously verified while establishing the DRF. Then, the other three holes are measured and verified to the DRF.

Straight forward…


Now, let’s examine the four hole datum “C” scenario.

If all four holes are used as datum “C” – the setup will be different. First datum’s A and B will be established as stated above. Then for each of the four holes a minimum of six areas of contact (three towards the top and three towards the bottom) will need be contacted and then a single derived axis from the four hole axis’s is established. Most modern CMM’s can handle this.

This single axis is then used to orient and lockdown the rotating the datum B axis/planes and thus lock all six degrees of freedom for the DRF. Finally after establishing our DRF the orientation and location of the four holes are measured and verified relative to the DRF.

For me – I think the definition is correct however the resulting DRF is different than the single hole Datum C DRF. Moreover, this definition may ~ confuse the metrologist and ultimately require extra time and work – depending on the technology capabilities of the CMM.

I don’t see the four hole datum specification as an improvement or advantageous
.

[Belanger]

Kelly, I would say that datum C would not be established from the four holes themselves. Datum C would be established from the maximum material boundary of the four holes. A fine point of distinction

What I've recommended is completely endorsed by the Y14.5 standard. It's unfortunate that it's hard to do with a CMM, but GD&T is based on function, not how easy it is to measure something.

That said, though, I believe there is a way for most CMMs to take actual data from a group of holes and virtually scoot the data around (sort of a best fit) within the MMB limits until the other position or profile tolerances comply.

I agree with your second point. Metrologists might get confused with such a specification. That too is unfortunate, but if we avoided certain callouts because people don't understand them or if people do whatever they want, the the entire GD&T system might as well be chucked out the window!

I'm teasing a little -- but hopefully there is a happy middle ground that allows the manufacturer to get the fullest tolerance and still have the part be verified properly.

[Kelly_Bramble]

I agree with your second point. Metrologists might get confused with such a specification. That too is unfortunate, but if we avoided certain callouts because people don't understand them or if people do whatever they want, the the entire GD&T system might as well be chucked out the window!

Yes, I agree with all of your points however I do think we as technical professionals should consider the simplest approach when we specify end-item requirements.

There are reasonable limits of technical knowledge expectations and that the “Learn it” statement can be elitist.

For example – let’s say that we have specified a requirement that uses the Pythagorean Theorem

Which, in simplest form is:
X² + y² = c²

If we give the end professional x and y it would not be unreasonable to expect them to understand algebra, manipulate the equation to derive c.

Or, a more academic approach we might express the proof of the Pythagorean Theorem which is

The slope of the line containing c is m = y/x.
and the line perpendicular to the line containing c at (x, y) is m = -x/y.
Now consider the set of all points (x, y) that are a fixed distance c from the origin where y > 0.
All possible points (x, y) in this scenario will lie on the solution curve to dy/dx = -x/y.

Finding a general solution to the differential equation dy/dx = -x/y we have
ydy = -xdx
∫ydy = -∫xdx
y²/2 = -x²/2 + D
x² + y² = E; E = 2D.

Now using an initial condition (x, y) = (c, 0) we have:
x² + y² = E
0² + c² = E
c² = E.

Do we expect our supporting technical professional to be proficient in Integral calculus? Is this a reasonable approach when specifying design requirements in the profit and schedule driven business world?

My point is, we in engineering and design should do all that we can to ensure manufacturing and quality are successful in creating fully functional end items and we should specify the simplest approach to get there. I'm not saying that should compromise design requirements just seek simplicity to get there..