As Long As We're Talking About Y14.5M

[swcalvert]

I, too, like Bob (400Magnum) am re-witting/creating a Drawing Standard for our company and I have always wanted to ask this question about Y14.5M. Some of the people I have talked to in the past have stated that ANSI Y14.5M requires you to dimension GD&T (true positioning) in the drawing mode. I argue that the standard also shows you other dimensioning schemes like Rectangle Coordinate Dimensioning (ref 1.9.1). What is everybodies thoughts. Do you True Position sheet metal parts?



Steve C

 

[dougr]

The title of ASME Y14.5-1994 is:



Dimensioning and Tolerancing



So it covers both Coordinate and Geometric Dimensioning and Tolerancing.



Position zones can be diametral or rectangular - depending on whether the diameter symbol is applied to the position tolerance.

 

[donha]

When should GD&T be used?

REF: Geo-Metrics Lowell W. Foster



1. When part features are critical to function or interchangeability;

2. When functional gaging techniques are desireable;

3. When datum references are desirable to ensure consistency between manufacturing and gaging operations;

4. When computerization techniques in design and manufacture are desireable;

5. When standard interpretation or tolerance is not already implied.



Rule #1 LIMITS OF SIZE applies (Y14.5m)

(2.7.1) Where only a tolerance of size is specified, the limits of size of an individual feature prescribe the extent to which variations in its geometric form, as well as size, are allowed.

(2.7.1.1) Variations of size

The actual size of an individual feature at any cross section shall be within the specified tolerance limit.



In other words, rectangular coordinate dimensioning geometrically controls a feature. In other words, GD&T is not a requirement.

 

[dougr]

Foster seems to be missing some things:



1) Features defined by coordinate dimensioning are always RFS.



Geometric dimensioning permits bonus tolerances depending on feature size making parts easier to produce.



Bonus tolerances with GD&T can often result in 100-200% additional tolerances resulting in less rework and better repeatability.



2) With coordinate dimensioning the tolerance zones are always rectangular.



With geometric dimensioning the tolerancing zones can be either diametral or rectangular.



Diametral zones result in 57% additional tolerance compared to rectangular zones.



3) With coordinate dimensioning the inspection criteria (datum-structure) must be assumed.



With geometric dimensioning it is always defined.





In a nutshell use GD&T because bonus tolerances are permissible and diametral tolerance zones can be used.

 

[Tunalover]

Dougr-

No, Mr. Foster isn't missing something. He was simply quoted incompletely and out of context. He is a leading authority on Y14.5M and drawing standards in general. Also, he has chaired the ASME/ANSI Y14.5 committee.



My experience (20yrs) with Y14.5M shows:



1. Because the method requires discipline and training for all involved, management generally dislikes the method.

2. All machine shops will tell you that they understand and practice the method but few really understand it.

3. When quoted by shops unfamiliar with GD&T (most of them), parts cost more when toleranced that way even if the tolerances are substantially looser.

4. The automotive industry has dramatically shown that millions of dollars in scrapped parts can be saved because, by virtue of the shape alone, an RFS circular tolerance zone has 56% more area than the equivalent unilateral tolerance zone. Of course when stated on an MMC basis, more savings are realized because a bigger hole affords an even larger tolerance (bonus tolerances).



As much as I like the method, I have found that American businesses lack the long view, maturity, and discipline necessary to take advantage of the method. In fact, many places I've worked don't even do incoming dimensional inspection, first article or otherwise. Result: mechanical engineers and designers (myself included) were put on the defense where they were forced to fix fit problems that didn't need fixing.



That's my say!

Tunalover

 

[dougr]

Seems to me purchasing depts should be smart enough to negotiate more favorable terms when 14.5 is invoked.



Why isn't GD&T taught in schools ???

 

[wcwyder]

The problem with using MMC to gain more usable parts is that these parts deviate more from the mean and thus create a smaller Cpk. This is totally againest the past trend of making no bad parts by keeping Cpks 1.33 or larger, and the current trend of 6 sigma which requires a Cpk of 2.0 or larger. Basically, using GD&T to define geometry is fine, but using MMC to gain usable parts is contrary to creating a tightly controlled process with a large Cpk.



I agree with Tunalover, most people realy don't understand the basic consepts of GD&T.



Walt

 

[dougr]

This is true.



It appears Quality Engineers don't understand GD&T either : )

 

[Nose Bleed]

First off, I'd like to note that it's not an ANSI standard - it's an ASME standard. ASME has apparently bought out ANSI on this particular standard.



However - being more familiar with the B31.3 standard - ASME has once again put forth another standard which must be completed with adendums... the adendums will continue until the next revision ( withmore addendums...)...



And with that in mind, I prefer that my QA is listening to the Prodigy - Fuel my fire - during his review....





The GT&d that I've found within the standard have fit my needs as a designer - with no frustrations (tongue in cheek)..



That being said, I prefer my designs to be checked with a CMM as opposed to QA analysis/visual inspection...

 

[Nose Bleed]

Speaking of... How does the standard govern Variable section swept cuts???

 

[Tunalover]

wcwyder-

You must be referring to some statistical term when you mention Cpk's. How about coming down to earth and joining normal folks like us? What's a Cpk?



How could bonus tolerances could be contrary to tightly-controlled processes? A tightly-controlled process is expensive. Production methods may vary in the product life cycle depending on production volume so why not look ahead and design/tolerance for them?



For example, John designs a widget with a production volume of <1000 pcs annually. He knows the parts will be machined so he freely invokes +/-.005 bilateral tolerances on all features. Bill, however, is forward looking; he uses GD&T to control the part function while keeping tolerances only as loose as he can tolerate. He uses loose tolerances because he concentrates on FUNCTION while anticipating higher production volumes that are not as forgiving as machining. For example, suppose demand for the product soars to >50,000 pcs annually. Bill and his company decide to diecast the widget. At that time, Bill and his company will be well-positioned to deal with die casting because as the die wears the size and positional tolerances will wander as the tool wears. Because all of his drawings were toleranced robustly from the start, there are very few drawing changes, Bill gets a promotion, and Bill starts tucking away a small fortune for his son's education.



A good habit is to design something as loose as possible while ignoring the production method. This is robust design that tolerates tool wear and other factors. Don't get me wrong though...GD&T shines when it comes to tight-tolerance designs too!



One last note: parts designed at MMC can be inspected very quickly using functional (ala go/no-go) gages. Inspection of bilateral tolerancing can only be done by CMM.



Tunalover

 

[Roger]

Guys:



Let's get our terminology straight. Geometric Dimensioning and Tolerancing is the overall standard. True Position is a subset thereof. And is NOT required to be using GD&T, e.g. a part with no holes.



Tunalover is quite correct. In addition, most designer/drafters overuse true position. By that, I mean they use tolerances that are WAY too tight. Many drawings I see have true position zones in the .005 range. Rediculous for most situations. Some are a little better at .010. Still very expensive. Think about it. If the circular tolerance zone is .010 that translates to about .0035 plus or minus. Reasonable? NOT!



Is it any wonder that machine shops charge more for True Positioned parts?



Oh, I forgot. I don't have a BSME. I know nothing.

 

[dougr]

most designer/drafters overuse true position. By that, I mean they use tolerances that are WAY too tight. Many drawings I see have true position zones in the .005 range.



This has nothing to do with GD&T, they use too tight tolerances regardless of GD&T.



If the circular tolerance zone is .010 that translates to about .0035 plus or minus.



The most common CYLINDRICAL diametral tolerance used with GD&T is 0.014. This equates to 0.010 (0.014*0.7071) in a rectangular tolerance zone.



In other words 0.014 on a DIAMETRAL CYLINDRICAL zone should be the yardstick for judging these tolerances.



True Position is a subset thereof.



Yes but it's a bloody great big one and by far the most important one.



btw the modern term for true position is just position



PS Some of the best Engineer's I've known were brought up on an apprenticeship system and don't have degrees - you should try to emulate them.



PPS GD&T is definitely the way to go despite what you think. Colleges need to start including this in their curriculums.

 

[doctordremel]

I was at PTC headquarters for a user group meeting and ran into a woman in the group that teaches Pro/E at a local college and she absolutely amazed me with her grasp of the whole Y14.5 code, my wife

 

[Nose Bleed]

I took a GD&T class two weeks ago - and the instructor mentioned that - that is the way they want to go (3d drawings, as opposed to ortho). But he did comment that it might not be implemented in our lifetime..



This was the first I heard of it.

 

[dougr]

The problem with 3D is you can't scale off of hard copies..

 

[Hacks]

doctordremel,



I don't think it was the latest edition of ASME Y14.5M-1994 the woman was refering to but rather ASME Y14.41-2003,

DIGITAL PRODUCT DEFINITION DATA PRACTICES.



Read the attached from CADCAMNet on 10-2-03.



dougr,



Scaling anything off of hard copies is a generally recognized poor practice. 14.41 speaks to using models only to define manufacturing and inspection requirements without the use of drawings.

 

[dougr]

generally recognized poor practice



Yes but it's still done..

 

[swcalvert]

Thanks for all your imput, as I stated earlier, I am in the process of re-writting our CAD drawing standard and I'll be sure that when I reference Y14.41 that I make sure that it is were applicable, because none of us are really sure what we're getting in to.



I'll keep this thread open for any more comments.



Steve C

 

[dougr]

Y14.41-2003, DIGITAL PRODUCT DEFINITION DATA PRACTICES



It looks like this standard will eliminate the practice of using driven dimensions.



According to survey taken on this site only a small percentage of modelers use driving dimensions in drawings.