TOLERANCE BLOCK 1

[kneeslider]

This is in regards to a response I saw from tunalover regarding tolerances:

UNLESS OTHERWISE SPECIFIED:

DIMENSIONS ARE IN INCHES.
TOLERANCES:
.X±.1
.XX±.01 (could be .XX±.015 too!)
.XXX±.005
.XXXX±.0005
ANGLES±1º

My question is "could be .XX±.015 too!". Is it permissible to use a 3 place decimal tolerance on a 2 place dimension?

The other question I had was if one can use ordinate dimensions with 2 and 3 place dimensions from the same 0 reference to control the tolerance.

 

[MechNorth]

Per Y14.5, for inch units, the number of decimal places in the tolerance must equal the number of decimal places in the dimension. So no, ".XX +/-.015" is not legal to the standard. For metric, however, they recognized that trailing zeros don't indicate precision (the assumption in the inch system), thus dimension and tolerance can have different number of decimal places.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[kneeslider]

I was hoping tunalover would repsond to this as he is the one who gave the example of the tolerance block. I know on the face of the drawing the number of decimal places for the tolerance must be the same as the dimension for inch units. I wanted to know why he thought it was ok to do this in the tolerance block as our Canadian operation has asked for this same tolerance.

 

[MechNorth]

Well, being "illegal" and being "common practice" are opposed aspects here. Most people don't know the rule, and wouldn't raise an eyebrow about it. If you need to keep things "clean", then you can put a note on the drawing or in a corporate addendum indicating that "An exception to Section ?? is made herein, and allows a different number of decimal places in the nominal dimension and assigned tolerance.", or such.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[CheckerHater]

Actually, this is kind of question I always wanted to ask:

According to ASME Y14.5-2009 Para. 2.3.2 (a) and (b) when we do DIRECT tolerancing, both dimension and tolerance shall be expressed with the same number of decimal places.
BUT
According to ASME Y14.5-2009 Para. 2.3.2 (d) in case of INDIRECT tolerancing using feature control frame, there is no requirement for dimension and tolerance to be expressed with the same number of decimal places.
SO
When we place untoleranced dimension somewhere on the face of the drawing, and corresponding tolerance is placed in the title block, this is in fact INDIRECT tolerancing (sort of).

Question: Where does it say that title block tolerance should follow clause (a) and not the clause (d) considering that toleransing based on the number of decimals is not in the book at all?

 

[KENAT]

Checkerhater, is that a change from -1994? In -94 2.3.2d for inch dims says "The basic dimension value is expressed with the same number of decimal places as the tolerance."

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[CheckerHater]

I double-checked, YES.

 

[pmarc]

CH,
I wouldn't stick to terms DIRECT or INDIRECT tolerancing in case of para. 2.3.2. Subparagraph d) merely describes the practice of tolerance expression if a geometrical tolerance is associated with BASIC DIMENSION.

Two examples:
#1. A hole defined by basic .500 dimension and a profile of surface .020 applied to the surface of that hole. One can put .020 or .02 or .0200 in profile FCF, because this is what the subpara. d) is talking about.
#2. A cylinder of diameter .098-1.002 plus straightness of its median line .010 applied to it. In that case, because the dimension which the straightness tolerance is associated with is not basic, the tolerance value should be expressed in three digits after decimal point format - so .010 and not .01 or .0100.

 

[fsincox]

Of course in metric it is different!

 

[CheckerHater]

Pmarc,

[ul]
[li]Well, Y14.5 does enough to muddy the water as it is.
Para. 2.1.1 describes direct limits and geometric tolerances as separate entities (clauses (a) and (b)), while Para. 2.2 states that they are both direct tolerances.[/li]
[/ul]

[ul]
[li]To me if tolerance is not attached directly to dimension and placed somewhere else on the drawing, it is still indirect.[/li]
[/ul]

[ul]
[li]Can you back your statement #2 with “direct” quote / reference to the standard?[/li]
[/ul]

 

[pmarc]

Yes, I agree the standard could be much clearer in case of paras. 2.1.1 & 2.2, and no, I am not able to back my statement #2 with any precise reference to the standard. This is how I see this whole stuff with number of decimals being equal or unequal for dimensions and associated tolerances. I should have probably added "in my opinion" in my previous post in order not to provoke you to ask me the question you asked.

 

[CheckerHater]

Pmarc,

“Provoked” is a strong word, just curious.

Sometimes some obscure places in standard need further clarification in textbooks or by other means. I was wondering if I missed something.

Considering that there is no current ASME/ANSI standard to deal with untoleranced dimension, it is designers’ right (and responsibility) to clarify it on the drawing.
Y14.5 gives you lot of freedom for that – see Para. 2.11 (c), (d), and (e) - so essentially whatever you put on the drawing IS the standard.

The way I see it, untoleranced dimension is what it is – sloppy design work.
On the other hand untoleranced dimension associated with some sort of note, table, etc with tolerances specified becomes “indirectly toleranced”, “nominal”, theoretically exact”, “basic” – you pick the name.
So I don’t see a big deal if they are treated one way or another.

But to think of it:
Is it legal to say “ALL UNTOLERANCED DIMS ARE +/- .01”? Should I then round up all of my untoleranced dimensions to match the note? What about design intent?
Also, I have nothing against matching number of decimals to the ACCURACY of the tolerance, but not to the number of digits.
“.XX UNTOLERANCED DIMS ARE +/- .01” is legal,
“.XX UNTOLERANCED DIMS ARE +/- .02” is legal,
“.XX UNTOLERANCED DIMS ARE +/- .015” is not legal? I don’t think so. What if .01 is to small and .02 is too big for me?

Another thing to consider: both ASME and ISO recognize dimensions and tolerances as absolute. There is no difference between .25, .250, and .2500.
So equal number of decimals requirement is purely cosmetic anyway.

All that said, in my opinion, if you clearly and unambiguously specify your rules for untoleranced dimensions on the drawing, it’s not a big deal if dimension specified in one corner of the drawing and tolerance placed in another corner of the drawing have different number of decimals.

 

[pmarc]

CH,
I fully agree with you.

I grew up in ISO (metric) world and to be honest I do not really understand the need of having equal number of decimals for a dimension and its tolerance in inch dimensioning. Your example with +/-.015 is exactly the reason why I have difficulties in grasping the true idea behind.

And like you said, dimension and tolerance values are absolute, so there is no difference whether print shows "1" or "1.0" or "1.0000" - it is always 1.

 

[fcsuper]

Fire (or educate) the drafter/engineering that added this requirement.

Matt Lorono, CSWP
Product Definition Specialist, DS SolidWorks Corp
Personal sites:
Lorono's SolidWorks Resources & SolidWorks Legion

 

[vonsteimel]

When the drawing gets to the shop, no one will know nor will they care that ASME says "the number of decimal places in the tolerance must equal the number of decimal places in the dimension"... If .XX = +/- .015 than that is what the shop will follow. I have done it many times.

One of my biggest shocks after leaving drafting school was how little the industry follows actual ASME standards....

 

[Belanger]

VonS -- that's all great until things get to the courtroom, where any deviation from the standard could be quite costly. Maybe not in this case of the number of digits after a decimal, but still...

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

http://www.gdtseminars.com