Question about Independency Principle 1

[pmarc]

Imagine that fig. 2-7 from Y14.5-2009 shows a pin instead of the rectangular block, and the flatness callout has been replaced by straightness FCF. Everything else stays the same (of course the diameter symbol is added to 10.7-10.8 dimension).

Knowing this, what is the minimum possible diameter of a perfect cylindrical boundary that the pin would never violate?

Thanks.

 

[CheckerHater]

Yes, better
No, they don’t control circularity

 

[Belanger]

Good stuff! I step out for one day and y'all run with some good comments. After the initial question, I then suspected that pmarc was referring to circularity rather than the longitudinal aspect.

Does CH's drawing of the 3-lobed part mean that the question has been answered? Or are we saying that ISO is a little ambiguous about this point?

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[pmarc]

What do you exactly mean by saying that "ISO is a little ambiguous about this point"?

 

[Belanger]

Well, I guess I don't really mean ambiguous. But here's what confused me: It sounded like the theory presented here was that ISO (or use of the I symbol in ASME) means that there is absolutely no control on circularity. I don't think that's true; if each diametrically opposed pair of points is within the size tolerance, there is still only so much deviation from one pair to the next.

Think of it this way: If we measure only 3 pairs of opposing points, then the circularity error could be awful. If we measure 10 pairs then the circularity is a little better. Now, keep increasing the number of pairs you measure ... independency still means that there are an infinite number of opposing-point pairs that must maintain size. If they are all right next to each other, there can't be too much of a jump in terms of their combined roundness. I hope that makes sense.

So is this the answer? -- The independency rule means that circularity is no longer limited within the size tolerance, but there still is a maximum limit to the circularity error.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[CheckerHater]

The reason they like to put lobed shapes in textbooks so much is that said shapes can be measured infinite number of times and return the same number.
I understand the frustration coming from ASME upbringing – the idea that diameter symbol demands perfect circle/cylinder, at least under some condition(s) is very comforting.
Entering the world where diameter is applied to part that may never be round is unsettling.
But guess what?
Just think of all shapes your “round” part may take when approaching LMC and you will realize that even under default envelope principle applied, roundness/circularity is not really controlled.
Imagine part DIA 10.000/.500 (E)
As perfect form is not really required at LMC, square .500 will perfectly fit both the tolerance (no less that .500 cross-section)and envelope (not exceeding 1.000 boundary)
So it’s all in your head, gentlemen

 

[Belanger]

I get that CH -- the lobe shape has often been used to illustrate that opposing points can give a false sense of security with respect to roundness. But do you really think that removing the envelope idea means there is absolutely no limit to the circularity error?


Also... a square of .500 x .500 is not really at LMC... that's a different story.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[CheckerHater]

Yes, square .500 x .500 is not LMC.
It is one of infinite number of intermediate shapes going from MMC to LMC that satisfy both tolerance an envelope requirement in my example.
It’s just it cannot have local measurement lesser than LMC size of .500.

No, I don’t think circularity error will be infinite. It’s just difficult to derive it mathematically.
But I am also curious about your opinion: if perfect round shape is only required at MMC, do you see any limits to how “ugly” shape may become while staying within the tolerance, and what is preventing it from distortion in absence of explicit circularity control?

 

[CheckerHater]

Here is my theory for Envelope requirement:
Circularity error is equal to size tolerance.
Now I will try to figure circularity error in Independence case. It may be larger.

 

[Belanger]

...if perfect round shape is only required at MMC, do you see any limits to how “ugly” shape may become while staying within the tolerance...

In the sense that we are discussing, perfect form is required not only at MMC, but at LMC. That is, only if we consider the cross-sectional aspect of a part -- forget the longitudinal aspect. In that case, perfect form is required at MMC and LMC. Think about it: if a diametrical shaft is truly at least material condition, it would have to be a perfect circle. If it has a ding going inward, then it violates size. If it has a spike going outward, then it's not at LMC!

The reason we say that perfect form is not required at LMC is that in the longitudinal direction the shaft could bend. It's still at LMC, and it's still within the envelope rule.

Anyway, I concur that with the independency concept invoked, circularity is not controlled by size. But there has to be some limit to the circularity error, and I suppose that it is more difficult to derive. So we're right back to pmarc's original question!

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[CheckerHater]

That’s very interesting
Could you quote where the standard says that perfect circularity is required at LMC?
Also, what about infinite number of shapes BETWEEN MMC and LMC?
Are all of them perfectly round as well?
Simple question: Part on my drawing. Will it pass the inspection? Simple Yes or No, please.

 

[Belanger]

I can't quote that from the standard. But I ask you to re-read my explanation: I was clearly limiting my statement to only the 2-D concept of a circle. If that still doesn't make sense, I will try to post a sketch later today.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[bxbzq]

size, actual local: the measured value of any individ-
ual distance at any cross section of a feature of size.

what does this mean to you? Which two points would you pick to measure odd shape?

 

[CheckerHater]

Bxbzq, If you are asking me, I already made this picture:

http://files.engineering.com/getfile.aspx?folder=9280bf62-df2d-4132-8235-1324a0671d9e&file=Draw1.JPG

The rule seems to be pretty clear:

From Para 27.1 (b):
“Where the actual local size of a regular feature of size has departed from MMC towards LMC, a local variation in form is allowed equal to the amount of such departure.”

So my part is departing from MMC and it still far away from perfect LMC cilcle.

Belanger, you put me in awkward position:
On one hand it’s impolite to insist that you answer my question.
On the other hand I only asked for simple “yes” or “no”.

 

[pmarc]

This is how I see the answer to my initial question and to the question asked by J-P about maximum possible circularity error when the independency principle is in charge. Feel free to comment and disagree.

http://files.engineering.com/getfil...57ca02acdea&file=circularity_independency.png

 

[Belanger]

Sorry CH -- yes, that part will pass inspection.

Now, my turn
Do you see how a part strictly at LMC (diametrically only; not axial-wise) has to be at perfect form?

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[CheckerHater]

Pmarc,
Where does it say that I am forbidden from measuring across the star?
Does “point-to-point” measurement mean “only between the points pmarc has chosen?
And as soon as I measure anything above 10.8, the part fails inspection.
I cannot believe you were serious.

Belanger,
If the drawing explicitly states that certain requirement applies at LMC, than perfect form is required.
That’s the letter of the standard.
In any other situation real part always “departs” from one to another.

 

[pmarc]

CH,
As for your measurement accross the star - this is one of the most serious shortcomings of Y14.5 standard. The definition of actual local size is really unclear. In ISO this is clearly standardized in a way that each line connecting two points must pass through the center of an associated circle.

So if it makes you feel any better, we can say that my picture follows ISO definition of actual 2-point local size.

 

[CheckerHater]

Pmarc,
You are going to have to explain it one more time, now with more words. Reference to the standard(s) may help as well

 

[pmarc]

ISO 14660-2:1999 is what you need - Figures 1 and 4 and associated paragraphs.

If you have no access to the standard, I can try to explain it, but to be honest nothing will be better than the two figures and the standard text.

 

[Belanger]

If the drawing explicitly states that certain requirement applies at LMC, than perfect form is required.
That’s the letter of the standard.

Arrrgh! You're not reading my posts, CH
Yes of course, if the shaft is at LMC then the LONGITUDINAL aspect is not required to have perfect form (straightness). But if you just consider one cross-section (which I clearly stated as the caveat), and that cross-section happens to be at LMC, tell me exactly how it could be out of round and still be called LMC?

Try it this way, using the MMC perfect-form rule: If an external circle is at MMC, you agree that it must have perfect form. Would you say that the circle could have a lobe extend outward? (No -- because it would exceed the size limit.)

Now, if that MMC circle had a ding inward, is that allowed? Yes or no?


John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com