Parallelism, location, dimensions, etc.

[bxbzq]

Hello guys,

My confustion originates from the training book from by Alex Krulikowski. On Fig. 7-15, a toleranced dimension (coordinate dimension?), highlighted (20.5, 20.8), is applied to a hole on a block, then a parallelism control is applied to the hole. See attached picture. While there are ways to apply parallelism control to control the parallelism of this hole, here are three I have in my mind:
Option 1, as the picture shows.
Option 2, locate the hole by a basic dimension, say 20.65, then apply a Position control, followed by a parallelism control to refine.
Option 3, locate the hole by a basic dimension, then apply a parallelism control. This is my first thought actually.

My first question: is the callout in option 3 legal? If it is legal, it would mean the parallelism can control location, which is not supported by the Standard. If not, what is the problem by doing it? Same question came to my mind when I looked back couple of pages in Alex's book, a parallelism control is applied to a surface that is one of the two opposed surfaces of a simple block. A tolerenced dimension (size dimension of FOS?) is applied to these two opposed surfaces. You can see Fig. 6-2 in the Standard, it is similar for this matter. So, what if the two opposed surfaces are controlled by basic dimension? Legal or not? Why?

Actually in option 1, my understanding is the parallelism control is a refinement to the toleranced dimension. This brings up another question. Section 2.7.4 in the Standard states:
"The limits of size do not control the orientation or location relationship between individual features. Features shown perpendicular, coaxial, or symmetrical to each other must be toleranced for location or orientation to avoid incomplete drawing requirements."

Interestingly the parallelism is not listed there, but included in subpara (a). Now back to fig. 6-2 in the Standard, the parallelism of the surface is obviously (to me) controlled by the size dimension. A parallelism control can be applied only if you want to refine the parallelism controlled by the size dimension. Something in the Standard seems to me not consistent.

I tried to search in the forum but did not find discussion specifically to this topic.

Please comment.

 

[pmarc]

bxbzq,
This picture from Alex's book is not something he should be proud of. Why? Because he uses directly toleranced dimension 20.5-20.8 to grasp a relationship between bottom face of the part and the axis of the hole. This should not happen and in my opinion is totally opposite to his statement shown in fig. 1-10 of the same book.

As for your other 2 options:
- option #2 is how this should be accomplished. Even if you are only interested in controlling parallelism of the axis, there has to be something that controls location of this hole too (and that would be the position tolerance), otherwise the geometrical characteristic of the feature is incomplete, thus violates one of the very fundamental GD&T rules;
- option #3 for the same reason as mentioned above is not a best choice, because you will not be controlling location of the hole. Keep in mind that basic dimension 20.65 would be meaningless for parallelism callout, because by nature parallelism does not control position.

 

[bxbzq]

pmarc,
It is interesting to compare the fig. 6-9 and fig. 6-10 in Standard. In fig. 6-9, no doubt there should be a basic dimension to locate the hole, the position and parallelism callouts are just fine. In fig. 6-10, however, we don't know if there should be a basic dimension or toleranced dimension. If it were a basic dimension, it would be same as option 3. If it were a toleranced dimension, it would be same as option 1. Either way, I'm just more confused.

 

[pmarc]

Lack of basic dimension on fig. 6-9 makes it confusing. I supposed it is not there because the intent of the figure was to interpret axis parallelism, for which this dimension - like I mentioned in my previous post - is useless.

For fig. 6-10, even if the basic dimension was there it would not change anything, so that is why it was probably not shown. Read para. 6.2 - it clearly states that orientation tolerances do not locate.

You have to remember that figures in the standard are quite often incomplete by intent (see para. 1.1.4). Unfortunately in this case the incompleteness brings nothing else than confusion.

 

[fsincox]

Option 3:
I do not like the term 'legal", it is certainly incomplete, if that makes it illegal then so be it.
Is this the ISO book?
Frank

 

[fsincox]

Option 1;
The old way!
Option 2:
The new way!

In industry both are common, 3 is not an option. There is an ISO standard that said it was an option, most of us say now is wrong and even the ISO people have confirmed it was a mistake, I believe?
Frank

 

[bxbzq]

Thanks to pmarc and Frank.
It is an asme book. There is 3rd edition on the way.

Seems my second question was not clearly addressed.
On one hand, the Standard says limits of size do not control the orientation and location.... Parallelism is one of the orientation, isn't it?
On the other hand, there is implied parallelism: unless otherwise specified, the parallelism of two surfaces shown as parallel on a drawing is controlled by the limits of the dimension between the surfaces. See fig. 6-2 and fig. 6-18 in Standard. In these examples, to me the parallelism callouts are refinement to the size tolerance.
Do I misunderstand the either one?

 

[fsincox]

Under ASME the limits of size DO control orientation, so your statement is wrong.
Frank

 

[fsincox]

The ISO does not exert this basic control this unless you invoke the "envelope principle" (symbol: circle “E”) which is automatic under ASME.
Frank

 

[pmarc]

bxbzq,
Imagine two pins shown on a print perpendicular to each other, having directly toleranced diameter dimensions assigned. For both features (or features of size, actually) upper limit values define envelopes that cannot be violated according to Rule #1. The intent of 2.7.4 is to say that those limits of size do not control mutual relationship between both envelopes. So despite the cylinders are shown perpendicular on a print, mutual perpendicualrity of the envelopes is not controlled.

The same can be transferred to a simple 2D rectangle described by width and height dimensions. Having only those two dimensions on the print says absolutely nothing about perpendicularity error allowed between side and bottom edge of the rectangle. In such case the rectangle can become parallelogram and the figure will still meet the print requirements. Does it help?

 

[Belanger]

I don't see anything wrong with Alex's picture, at least in the way you guys are seeing it.

It's not illegal to have limit dimensions control the location of a hole. It's not wise, but it's legal (earlier in the same book, he emphasizes that coordinate tolerancing is to be used only for size, chamfer, and radius, so he violates his own suggestion, but not the standard).

And parallelism never uses a basic dimension, so that option is out.

One thing about Alex's picture, however, is that there is a diameter symbol, even though the only relationship to the datum is up/down. If a side wall formed a secondary datum, then the diameter symbol would make more sense to me. Maybe it still as the effect of controlling axis straightness?

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

http://www.gdtseminars.com

 

[pmarc]

J-P,
If he wants to use fixed gage for axis parallelism verification, there has to be diameter symbol. Otherwise such gage could not be used.

 

[powerhound]

I don't think Krulikowski is using the example as a recommendation of how to locate the hole, I think he is only providing an example of what it means to use parallelism to refine the orientation of a feature of size. The better way to locate the hole is to use position and then refine with parallelism if necessary. This, however, introduces elements that could potentially detract from the point that the illustration is trying to make. To put it in context, this chapter of his textbook deals strictly with orientation and this page is from the section on parallelism and how to apply it to a surface and to a feature of size. It is tantamount to an "incomplete" drawing as is shown numerous times in the standard, although it's not really incomplete.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[fsincox]

"Under ASME the limits of size DO control orientation, so your statement is wrong."

I am sorry, I was wrong: “orientation” should have been “parallelism”. At work I tend to look at this stuff too quickly.

 

[Belanger]

Right-o, pmarc. Thnks!

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

http://www.gdtseminars.com

 

[bxbzq]

pmarc,
Correct me if I'm wrong.
There is implied 90° angle, tolerances in title block or general notes control it, not the limits of size. This is how the perpendicularity of the two pins in you example, and the squareness of your 2D rectangle, are controlled. There is also implied 90° and 0° basic angles, go to the geometric control for tolerance. Either one, limits of size do not play a role. I think I get more on perpendicularity than on parallelism.
Don't know why standard does not say implied 0° angle though.

Back to parallelism. I read the section 2.7.4 again just a few minutes ago. The first para says orientation, then go to perpendicularity. No parallelism, no angularity. @##$@$%wtf+^&^U%... I thought when it says orientation, it would apply to perpendicularity, parallelism and angularity. wow, now I realize it is not true. Just perpendicularity applied!

Frank,
I felt I missed some "ABC" things. This is it, isn't it?

Ah, wait a second, I almost forget this. Looking at fig. 8-27 in 09' standard, learned from a thread a month ago, the size limit controls location of the top surface. Ooops...

 

[fsincox]

Bxbzq,
Basically, it is all important though because it builds upon itself like mathematics.
Under ASME rules and the “envelopes principle” size does control parallelism; so generally a parallelism callout value should be a refinement of the size tolerance. The other orientation tolerances are commonly controlled by some general tolerances, as you suspected, but they must be stated/referenced somewhere.
Frank

 

[axym]

"One thing about Alex's picture, however, is that there is a diameter symbol, even though the only relationship to the datum is up/down. If a side wall formed a secondary datum, then the diameter symbol would make more sense to me. Maybe it still as the effect of controlling axis straightness?"

J-P,

I still say that the presence or absence of datum feature references doesn't influence the shape of the tolerance zone. It's a cylindrical zone that can freely translate, not a parallel-plane zone. This is the same issue as we discussed in Frank's thread about the sphere - the effects of datum feature shift don't change the shape of the tolerznce zone ((IMHO). But I think we agreed to disagree on that one.

I'm not letting you off the hook on the use of the term "axis straightness" though ;^)

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fsincox]

J-P,
Sorry, I thought you had agreed that we could use a cylindrical zone despite what the actual stated framework supported??

Wow, I am not too crazy about this, "only if I am going to gage it" stuff. How am I supposed to know how they are going to check it?
Frank

 

[Belanger]

I never disagreed with the cylindrical shape of the tolerance zone -- you guys got me all wrong.

The shape of this zone is clearly a cylinder; I simply mused about the value of such if the sideways direction isn't controlled. But I like pmarc's answer.

Evan ... "axis straightness" ... huh? Maybe referring to my thread about the derived median line stuff?

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

http://www.gdtseminars.com