Tolerances and decimal places 2

[JV1]

I would like to know what your opinion is regarding the number of decimal places of form tolerances. ASME Y14.5 says that inch dimensions and their tolerances should have the same number of decimal places e.g. .500±.125 or .3125±.0010. Now, if we add a form tolerance, say circularity. Should the form tolerance have the same number of decimal places? For instance, if the dimension is .3125 +.000/-.0030, should the circularity be .0010 or .001?

 

[ewh]

Yes, it should contain the same number of decimal places.
See para 2.3.2(d) of ASME Y14.5M-1994.

 

[ctopher]

ewh is correct.
Also, check your CAD settings, some are set to drop the trailing zero.

Chris
SolidWorks 07 4.0/PDMWorks 07
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[looslib]

Trailing zeros must be dropped on metric dimensioned drawings. If the you do both, english and metric, then you will need to adjust your SW settings.


"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[JV1]

EWH, thank you very much for your responce. Unfortunately, I think the reason you gave does not apply in this case. Paragraph 2.3.2(d) of ASME Y14.5M-1994 applies to toleraces associated with a basic dimensions. In my example, the feature is a cylinder and it is the primary datum feature.

 

[dingy2]

Yes - in inches it is the same number of decimal places

Dave D.

http://www.qmsi.ca

 

[JV1]

Why?

 

[ctopher]

Yes, metric dims drop trailing zeros.


Chris
SolidWorks 07 4.0/PDMWorks 07
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[ewh]

The example shown may be of a locational tolerance, but the para does not specify which type of tolerance it applies to, only that the "basic dimension value is expressed with the same number of decimal places as the tolerance."

If you interpret this as meaning that tolerances other than positional are exempt because they are not specifically referred to, you are going to have quite a few problems applying and understanding this standard.
Please refer to para 1.1.4.

 

[KENAT]

I believe ewh is correct for decimal inch.

" For instance, if the dimension is .3125 +.000/-.0030, should the circularity be .0010 or .001? "

My understanding of the standard is that it should be:

.3125 +.0000/-.0030 cirulariy .0010

The common element on all 4 cases in 2.3.2 is that the tolerance has the same number of figures as the dimension.



KENAT, probably the least qualified checker you'll ever meet...

 

[ewh]

Yes, I was only refering to decimal inches...

 

[JV1]

What happens if we change the problem a little bit. Let's say the feature is the primary machining feature in a casting where the location of the feature is not too important. This primary machining feature is located from datum targets using basic dimensions. So we have the following:

Hole size: .3125 +.000/-.0030 inches,
Should the circularity be .0010 or .001? As state above you agree it should me .0010. Right?

Now, if the TP of the hole is .1 and the basic dimensions locating the feature from two sides of the part are 1.0 and 2.0 respectively. As they are (.1, 1.0, 2.0), they comply to para 2.3.2(d) of ASME Y14.5M-1994. Should these be also four decimal places (TP = .1000 and 1.0000 and 2.0000)?

 

[ctopher]

Hole size: ".3125 +.0000/-.0030 inches",
yes, circularity should be .0010

Chris
SolidWorks 07 4.0/PDMWorks 07
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[ringman]

The .3125, which isa a fractional tool size drives us to use a 4 place dimension and tolerance. However, there is no gain and I do not see the requirement for a 4 place decimal for a circularity tolerance in this case. The trailing zero is of NO VALUE.

 

[KENAT]

"The trailing zero is of NO VALUE." However I believe the standard strongly implies, if not explicitly states in this case, that you should have it.

KENAT, probably the least qualified checker you'll ever meet...

 

[ctopher]

I agree it seems to be no value, but I also believe in the standards.
Reduce errors, stay consistent.

Chris
SolidWorks 07 4.0/PDMWorks 07
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[ewh]

I agree with KENAT. If there is no gain, a solution is to round the size tolerance to three decimal places, not ignore the standard.

As for the OP's second clarification, I think that if the positional tolerance can be stated using one decimal place, the associated dimensions could also be one decimal place. The decimal places of the locating dimensions would still be the same number as that of the related tolerance.

 

[JV1]

So what happend to "The common element on all 4 cases in 2.3.2 is that the tolerance has the same number of figures as the dimension"?

Let's review:

We agree that the size dimension and tolerance should have the same number of decimal places: 3125 +.0000/-.0030 (para 2.3.2(b) of ASME Y14.5M-1994).

We agree that the "tolerance can be stated using one decimal place, the associated dimensions could also be one decimal place. The decimal places of the locating dimensions would still be the same number as that of the related tolerance.". Thus, TP = .1 and basic dimension = 1.0, 2.0 is correct (para 2.3.2(d) of ASME Y14.5M-1994).

I would say that form tolerances should have the same decimal places as the size dimension of the FOS or non-FOS they define. I will call this rule "good practice" but it is not in the standard (circularity of .0010). Do you agree?

 

[ewh]

I agree that form tolerances should have the same decimal places as the size dimension of the FOS.

I have learned to keep a somewhat open mind when it comes to understanding this standard. Different situations call for different solutions and the standard does not cover every possible situation, but does define the basic groundwork on how to approach them.

 

[KENAT]

"I would say that form tolerances should have the same decimal places as the size dimension of the FOS or non-FOS they define. I will call this rule "good practice" but it is not in the standard (circularity of .0010). Do you agree?"

JV1 I would go a little further and say it was implied by the standard but not explicitly stated.

KENAT, probably the least qualified checker you'll ever meet...