All tolerances are non-accumulative 1

[genes]

I am performing a loop stackup tolerance study on a sheetmetal assembly that has some assembly issues and have encountered an interpretation conflict. The note in the subject line is on every drawing. Before the firestorm of criticism begins, I was a student of Alex Krulikowski's in Michigan so I know what needs to be done with these drawings. That refinement isn't going to happen at this company. My question is, is this note a crude attempt at basic dimensioning? I was instructed, in no uncertain terms, that the hole you examine has +/-.030 for location and is not influenced by the previous hole to hole dimensional location tolerances. One can see the problem with this logic in a stackup if I have to disregard the previous location tolerances. Has anyone dealt with this kind of situation or could recomend a good shrink? Perhaps a source noting how wrong this drawing note is?

 

[KENAT]

It does sound like their attempt to imply a kind of basic position/dimensioning. Whether it would hold up in a court of law is another matter, which is part of the reason why the standards have developed like they are. I just did a quick google and couldn't find a reference but I believe at least one has been posted on this site before.

Not sure about a good shrink but I've had vaguely similar issues before with drawings referencing iso2768 thread1103-196260 thread1103-197786.

Posting guidelines faq731-376

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

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

 

[ewh]

I have seen this used properly before with an overall dimension "(6X 2.750=)16.500" and a single reference dimension "(2.750)", which does denote a non-cumulative condition between the holes. As shown however, there is nothing to indicate that it is non-cumulative.
I would also question using "TYP" and the dimensioning of a 90° angle.
I may have been working in the sticks, but I have never heard of Alex Krulikowski.

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]

 

[axym]

So they're saying that, for example, the tolerance between the second-from-left hole and the far-right hole would be +/- .030 even though the individual tolerances would stack up to +/- .150. I'm not sure that it's an attempt at basic dimensioning, it seems more like an extension of the general tolerance concept. Any given feature needs to be located within +/- .030 relative to any other feature. I agree that you're going to need psychiatric help after attempting a stackup study on that.

There are a couple of other odd things about this drawing. The block tolerance for two-place decimals is +/- .03 and for three-place decimals it's +/- .030. Hopefully these weren't intended to have different meanings. Also, there is an explicit 90 degree angle shown - I haven't seen that in a while.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fcsuper]

It seems like the whole company (or one particular individual in a powerful position) is confused about how dimensions and tolerances work.

The example you show is not an attempt at basic dims. Although basic dims would solve the issues with this drawing. Axym's interpretation seems to be accurate.

Essentially, it seems they are trying to employ a very crude positional tolerancing scheme via the use of linear dims/tols. From this interpretation, and how the drawing is dimensioned, each hole has multiple tolerance zones (at least 9 from a quick count) within which it much fit all. A simple positional tolerance only has one tolerance zone. This attempt to make a simplified drawing (by not using GD&t) actually makes it much more complex.

Bottom line, there is no such thing as "non-accumulative" tolerances in any standard. If this is the method they wish to use, then they should define what this means on the drawing. In fact, there's a lot of defining that will be necessary, as I see no reference to any drafting standards.





Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[MartinSr00]

Even with such a note, what does it actually mean? One of the jobs of GD&T is to note engineering intent (see fundamental rules from the 2009 standard, for example). That intent would be described, for example, in a location feature control frames describing the associative datum features. With all that missing, you're forced to guess.

How are you even to reckon the tolerance? perpendicular to the top face of the part? The bottom? The mid-plane? Exactly where.

If you don't use The Standard, to be clear, you need to adopt another one. Merely putting a note of non-accumulation doesn't do this job.

If your organization wants the leave things fuzzy for the sake of simplicity, then so be it. But fuzzy drawings, are, well, fuzzy. When there's a fit problem, fuzzy drawings are a down payment on a technically unresolvable dispute.

 

[genes]

Thanks to all who have responded. FCSUPER made a very good observation about the culture of the company. As for Alex Krulikowski, I was in his class for about a year learning G.D.&T. for my job years ago at BorgWarner, brilliant educator and has many good informative publications. I tried to introduce G.D.&T. here and was nearly burned at the stake. This time around, it's how to do coordinate dimensioning tolerance stacks, which many of you know is very difficult compared to positional methods. Compound the goal with drawing notes as in the subject line, and you get attitude and a sense of despair. Thanks to all.
GeneS.

 

[KENAT]

genes, you have my sympathy, my issues with this got ugly, I thought I might lose my job to it, instead they demoted and then 'let go' the other guy. Not sure it was directly related but I suspect it was a factor.

Posting guidelines faq731-376

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

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

 

[drawoh]

genes,

It sounds to me like you need to win an argument with someone.

If datums were clearly specified on the drawing, the concept of non-accumulative tolerances might mean something. In the absence of them, I would want to know what feature they were not accumulating from. Tolerance will accumulate from all the other features.

Could you take one drawing and re-do it with proper datums and GD&T FCFs? This gives you a chance to demonstrate how your features are controlled unambiguously.

JHG

 

[KENAT]

Having been part of a team introducing adherance to standards to a company, I can tell you it will most likely get ugly, and you wont win many friends.

Don't expect logic to win the day.

Posting guidelines faq731-376

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

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

 

[genes]

Kenat, I can relate to your post. The same nearly happened to me. It was not as extreme as your experience, just demoralizing. I can not comprehend how such a usefull tool as G.D.&T. becomes so maligned. Most here have no engineering or drafting training or education. As a final statement to an ASME/ANSI discussion I was told once by a manager, "we bend metal here, that's it". The manufacturing industry is a lot less enlightened than we think it is sometimes.

 

[fcsuper]

Usually, resistance to GD&T comes from fear and lack of understanding its intent.

Many vendors still use the sight of a FCF as an excuse to jack up the price on their quotes. You got to call them on it and let them know that you know what the FCF means and that you know they know as well.

Best of luck!!!!

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[MechEng2005]

It appears to me that the dimensions should then be baseline type. Then, each hole will be located from the same location, and no stack-up applies. Using the "2.750 TYP" indicates that the holes should be located with respect to each other, where the note seems to imply this is not the case.

Of course, if baseline dimensions were used the 2.750 dimension could end up being 2.810 to 2.790 and still be acceptable.

The "simpliest" solution to me would be to use baseline dimensions and adjust the tolerance to achieve an acceptable amount of spacing for the "2.750" dimenions (i.e. maybe the first hole is located 1.410 +/-.01 from the end, the second hole is 4.160 +/- .010 etc...)

-- MechEng2005

 

[ModulusCT]

My boss has told me on several occasions that we don't use GD&T often because our parts aren't that critical.

*smacks forehead*

 

[ewh]

Thanks for straightening me out, fcsuper. It was in a standard that you could designate non-cumulative dimensions as in my example, but it was in a company standard, not anything otherwise recognized by industry.


"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]

 

[PeterStock]

If that company standard is referenced on the drawing then it can be considered part of the drawing and define the meaning of "non-accumulative".

Peter Stockhausen
Senior Design Analyst (Checker)
Infotech Aerospace Services

http://www.infotechpr.net

 

[KENAT]

Although whether is is well enough defined to stand up in a court of law may be another matter.

Posting guidelines faq731-376

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

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

 

[ewh]

It was so referenced, but has no bearing on applying current standards to the OP.
As for standing up in a court of law, the company has long since faded from existence, so it is now a moot point.

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]

 

[NomLaser]

Sounds like they are just being lazy and trying to keep from having to calculate their stack-up and using proper tolerancing.

 

[MartinSr00]

I worked at a shop with such a culture.

What you always hear is "it's just not necessary". When they say that, it means that they don't want to think very hard.

Remember one of the most important fundamental rules of the 14.5 Standard -- "Drawings shall be subject to one interpretation".

It would be nice if such people simply said, "I hate to think very hard. Let's just fight it out after it becomes a problem." At least the virtue of honesty would shine through.

Now that's an argument I can understand (even if I don't agree with it).