precision in tolerancing

[paul16510]

ref. thread thread1103-129116

A question was posed to me that i would like to pose the community. It has to do with using two and three place decimals.
The dimension is 25.50 +.00, -.02
it has been rewritten as:
25.50+.000-.020
25.50+.00-.020
25.50+.000-.02

The question is are these all the same? I say no they are not and here is my reasoning.
25.50+.00-.020-------25.501 is acceptable (rounding convention), 25.479 is not acceptable the smallest value can be 25.480.
25.50+.000-.020------25.501 is not acceptable, 25.479 is also not acceptable
25.50+.00-.02---------25.501 & 25.479 are both acceptable conventional rounding
25.500+.000-.020-----25.501 & 25.479 are not acceptable.
What does the audience think?

 

[ctopher]

They are all the same.
25.50-.02 does not = 25.479 on a drawing.
If a part is machined to this, it's out of print.

Chris
SolidWorks 11
ctopher's home
SolidWorks Legion

 

[pmarc]

Per paragraph 2.4 of ASME Y14.5 all dimensional limits are absolute. That being said:
25.0 means 25.0............0
24.8 means 24.8............0
Everything that is outside the limits is out of tolerance and does not meet print requirements.

 

[Belanger]

Paul, I'm presuming you're using English units (inches). According to Y14.5, only the original spec is correct. Those which are rewritten as:
25.50+.000-.020
25.50+.00-.020
25.50+.000-.02

are not correct per paragraph 2.3.2(a):
"Where unilateral tolerancing is used and either the plus or minus value is nil, its dimension shall be expressed with the same number of decimal places, and the appropriate plus or minus sign."

That's an easy way to answer your colleagues' question

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

http://www.gdtseminars.com

 

[paul16510]

Thank you I was looking for the ASME reference. I understand it but I do not necessarily agree with it. I'll sleep on it and agree by morning.

Thank you again and Have wonderful weekend.

 

[paul16510]

Thank you Belanger. and yes I am using English units
I was unable to to obtain that spec this afternoon.
I was looking at the problem while keeping in mind significant figures and the level of precision (2 place decimal) provided by the number (25.50).

 

[israelkk]

As far as I know the number of digits also dictates the accuracy of the measuring machine. For example 25.1 can be measures by a regular caliper (0.05 accuracy) but 25.10 needs to be measures by a micrometer with can measure with greater accuracy (0.005 accuracy).