Decimal places on Drawing (Print) Callouts

[goodguyj]

What is the ruling on decimal places on drawings and what number of decimals do you measure them to?

If a diameter is 35.0mm +/-0.1

Measured with digital micrometer 35.11 on display, is this a failed part?

My colleague is claiming that is is not due to the drawing only states 1 decimal point.

Please help

 

[Jessterish]

I would tend to agree with your colleague, as the engineer should understand significant digits and act accordingly. Though it also depends on what kind of tolerances the reciprocal part has. if this is a hole for a part that has a tolerance in the opposite direction of .1 and both are on the wrong side, it won't fit.

EDIT: I stand corrected, apparently we are supposed to use infinite sig digits as long as the resolution of our measuring equipment is capable. Noted for further reference.

 

[quantim]

ASME Y14.5-2009, the GD&T specification, in section 2.4, reports that there is an infinite number of zeros after the last shown digit on a drawing.

"All limits are absolute. Dimensional limits, regardless of the number of decimal places, are used as if they were continued with zeros."

The examples includes "12.2 means 12.20000000"

Therefore, in your example, the part is non-compliant with the drawing.

This makes sense from a practical standpoint since this will assure a slip fit will remain a slip fit, and an interference fit will maintain an interference fit.

It is highly recommended that GD&T is used on ALL drawings to eliminate ambiguity, reduce costs since tolerances can be larger, and assure parts will always fit together.

 

[ScottK]

in my world if a dim is 35.0mm +- 0.1 that means if you are using a gauge with higher resolution anything greater than 35.1 or less than 34.9 is out of spec.

35.11 is higher than 35.1. Rounding can cause problems... ask your engineering what he would consider 35.18mm? Is that in or out? If you round up it is, but if you truncate the last digit it's not....

mostly I have specs that may look like 1.000" +-0 .005".
most of our calipers have a resolution of .0005".
If the dimension measured is 1.0055" it's out and gets an NCR... in the end the decision may be use-as-is, but there should be an action to make sure next time the parts are closer to nominal.

 

[Eredhel]

We don't get to round tolerance numbers in whatever direction we want , it's out of tolerance. But are you able to talk to your customer to get one time acceptance? Is it close enough they might not mind?

 

[Scanton]

I work in manufacturing and we make parts for many industries including automotive and I would agree 35.1 is 35.100000000000 etc, however that said if you talk to your customer 35.11 may not be a problem for them and they can give you a one time waiver for the quantity you have produced.

If it is a problem for your customer then the only options are rework (if possible and/or acceptable) or scrap.

My conversation with the person who made them would be along the lines of "you had a 200 micron limit, why did you make the 10 microns over size?"

 

[David-D]

As mentioned above ASME Y14.5 states unequivocally that you treat it as if you had an infinite number of zeros and the part is nonconforming.

Additionally, you undertake risk by having less accurate measurement equipment because you have greayer risk pf potentially reject good product or accept bad product due to measurement uncertainty/error. Many people use the 10% rule (gage must be atleast an order of magitude more accurate than the tolerance) while others implement pesimistic tolerancing of 540.3 where you reduce the product acceptance tolerance by the accuracy of the gage to ensure you don't accept nonconformig product.

David

 

[goodguyj]

Thank you all for the information and advice. I have worked in manufacturing for many years in the United Kingdom, then 4 years ago I moved to Australia. I was starting to doubt myself when questioned. The parts have been approved and will be processed under concession.
Once again thanks everyone.