Rounding of Tolerances 1

[randy64]

This is a question about rounding tolerances. It is a product of dual dimensioning. I know the evils of dual dimensioning, but our customer does it, and so it is what it is.

I searched the archives but have not seen this particular question as it applies to tolerances.

Question: When converting the inch tolerance in the FCF of .014 it gets us 0.3556. Our rounding standards state that a tolerance of .014 will get rounded to 2 places for the metric. I am arguing that the metric number in the FCF should be 0.35, due to the principle of never adding tolerance (could allow a too-big part that won't fit/go together). However, we have another general standard that defines how to round numbers. It's the usual stuff about 0-4 rounding down, 6-9 rounding up and rounding 5 depending on even/odd.

Bottom line, I am meeting resistance by folks that are blindly following the rounding standard i.e. .199 rounded to .20, versus my belief that a tolerance should never be rounded up i.e. .199 rounded to .19.

Is there something in AMSE that addresses this that will back up my position (assuming I'm correct. If I'm not I'm sure you all will let me know)?

Thanks.

 

[ctopher]

Did a quick search here.
thread1103-225708

Chris, CSWA
SolidWorks 14
ctopher's home
SolidWorks Legion

 

[randy64]

ctopher, I saw that thread in my search, but it doesn't address the rounding of tolerances. Unless your saying that tolerances should be rounded the same as any number/measurement?

 

[SeasonLee]

Try to see this Tec-Ease tip (Rounding Numbers):

http://www.dimcax.com/gdt_web/september-05.htm

Season

 

[randy64]

Thanks, SeasonLee, that confirms what I thought.

In the tip it says, "The ANSI/IEEE 268 standard, referred to in Y14.5..." I looked in Y14.5 (we use 1994) but could not find ANSI/IEEE 268 referred to, at least not in 1.2 REFERENCES. Anyone know where that might be? I'd like to have unrefutable documentation when discussing this.

 

[randy64]

Found it. Section 1.6.4 Conversion and Rounding of Linear Units.

 

[SeasonLee]

Don't miss the link recommended by Chris, the NASA Engineering Drawing Standards Manual, section 10.2.2.2 Unit Rounding, its a good reference.

http://mscweb.gsfc.nasa.gov/543web/files/GSFC-X-673-64-1F.pdf

Season

 

[randy64]

I'm sure it's a good reference but will be met with the following response by my colleagues: "That's not our standard! Our customer is [large aviation company in the midwest], not NASA!"

 

[powerhound]

I'm with you on this one but everyone shouls be aware that if a part falls outside of the rounded zone but still within the 4 place decimal zone, then the part is still good. There should be some sort of system in place to ensure that happens.

John Acosta, GDTP S-0731
Engineering Technician
Inventor 2013
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[ajack1]

Unless I am missing something obvious whilst rounding up can take something over top limit rounding down could also take it under bottom limit.

To my understanding there is no way of applying exactly the same limits in both imperial and metric whilst using any known standards?

The best solution is to never dual dimension but failing that the limits must be tighter in the secondary system.

 

[randy64]

ajack1, I agree with you, assuming you're talking about limit dimensions. My specific example was about the tolerance we put in the FCF, which means always rounding down since it is an upper limit.

 

[MintJulep]

"In the event of conflict between inch and metric dimensions inch dimensions (and tolerances) shall have precedence."

 

[DeSimulacra]

I agree with MintJulep, if inches are the primary. I was taught that the secondary dimension be it metric or English is there for reference only and that the primary dimension is what tolerance is derived from. How else could you determine what the tolerance is if the English and metric have different decimal places?
For reference I have just finished converting customer drawings with english being primary and default rounded to 2 places, some english dimensions were at one place and a few at three. The secondary was metric and every dimension was rounded to 3 places. The draftsman must have been taught as I have been.

 

[ajack1]

If you cannot manufacture to the secondary set of dimensions what is the point in putting them on the drawing?

 

[randy64]

As I said at the top, I'm not looking to debate the merits or correctness of dual dimensioning. My customer wants it and so I have to work within those parameters.

DeSimulacra, you said, "I agree with MintJulep, if inches are the primary. I was taught that the secondary dimension be it metric or English is there for reference only and that the primary dimension is what tolerance is derived from."

On some of our drawings, where the metric is primary, we design to the inch and derive our tolerances from the inch and then convert to metric. I know, it's not proper and probably wrong, but when the 800 pound gorilla says to do it that way, you do it.

 

[ajack1]

If you want to be able to manufacture to them and not have rejected parts then the secondary set of dimensions / tolerances need to be tighter. If they are just for some kind of reference and serve no real purpose then it doesn't really matter what you do.

Fortunately unless you happen to work for any American companies this is not a problem for the rest of the world.

 

[JNieman]

"The rest of the world" still dual-dimensions when dealing with countries that may not use metric as often. It's not isolated to the USA.

_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5

 

[CheckerHater]

On some of our drawings, where the metric is primary, we design to the inch and derive our tolerances from the inch and then convert to metric. I know, it's not proper and probably wrong, but when the 800 pound gorilla says to do it that way, you do it.

No wonder 800-pounds gorillas eventually went extinct.

 

[ewh]

Nope, still a few running around!

“Know the rules well, so you can break them effectively.”
-Dalai Lama XIV

 

[DeSimulacra]

Randy
You said"
On some of our drawings, where the metric is primary, we design to the inch and derive our tolerances from the inch and then convert to metric. I know, it's not proper and probably wrong, but when the 800 pound gorilla says to do it that way, you do it."

All that is fine, just remember, If you dimension in Inches and secondary in metric the default tolerance block will/should be in inches. The opposite will hold true if primary is metric. What you design in holds no relevance.
Another way to put it is why would you make your primary units inches and then tolerance in metric?

DeSim