Metric tolerances in title block 4

[JJAV1983]

Hi,

I recently started working for a new company, I noticed that the default metric title block we are using only has the specifie tolerance as follows:

TOLERANCE +/- 0.1 (mm)

All current and past drawings are dimensioned with 2 decimal places. So for example right now if we dimensioned the width of a plate as 19.10 mm, with the only reference we have in title block (Tolerance +/- 0.1) this width could be made at 19.00 mm or 19.20 mm. But what would happened if the width is 1900.00 mm and we don't need a tolerance as tight as +/- 0.1.
There is no specification for decimal places whatsoever, the only thing it says is: TOLERANCE +/- 0.1
We design from PCBs to low, medium and line voltage lighting systems and every single aspect of the fixture (housings, electrical boxes, brackets, etc). I really think we need to make some modifications to the title block, as the specification is real vague.

I was thinking something like this:

UNLESS OTHERWISE SPECIFIED:
X +/- 0.5
X.X +/- 0.3
X.XX +/- 0.13

Thank you for your advice.

JJAV 1983

 

[pmarc]

JJAV1983,
Let me ask you one question:

There is a statement in appendix to ISO 2768 (and probably in B4.3) that says:
"These advantages [of using ISO 2768] are fully obtained only when there is sufficient reliability that the general tolerantes will not be exceeded, i.e. when the customary Workshop accuracy of the particular Workshop is equal to or finer than the general tolerantes indicated in the drawing. The Workshop should, therefore
- find out by measurements what its customary Workshop accuracy is;
- accept only those drawings having general tolerantes equal to or greater than its customary Workshop accuracy;
- check by sampling that its customary Workshop accuracy does not deteriorate."

Have you had a chance to investigate that these conditions are/can be met in your new company?

--------------

DiegoLGraves,

The newest edition of ISO 8015:2011, "GPS - Fundamentals - Concepts, principles and rules" states in paragraph 4.4 entitled "Workpiece functional level" that: "It is assumed for interpretation that the workpiece functions 100 % within the tolerance limits and 0 % outside the tolerance limits." How would you refer to this sentence?

--------------

All,
Please don't get me wrong. I do not think ISO 2768 (or ANSI B4.3), at least part 1, is pure evil. That is not true. All I am trying to say is that serious and thorough analysis have to be done before such record in drawing general notes appears. Otherwise it can really cause more harm than benefits.

I do not even want to start again a discussion whether the tolerances specified in this standard back in late 80's are realistic in 2010's. I think we all could write a book on this topic based on our experiences or spend countless hours on debating what is right and wrong. The bottom line is - don't do it just like that. It is not that easy as it seems to be.

This is just my opinion.

 

[fsincox]

Mr. Graves,
I have been wondering how this works for people who actually use it. That same statement is not very popular with many on this board, myself NOT included.
CH,
Are we sure it is not popular outside of our own little world, here? Really GD&T (ASME or ISO) is not popular outside of here for the most part!
Frank

 

[JJAV1983]

pmarc

At the headquarters we just design the fixtures, most of our parts are being made overseas. Whenever the manufacturer send a quotation to us it says TOLERANCE: ISO2768-m, therefore they can meet the requirements no problem.

I really think that wheter referring to ISO2768 or ANSI B4.3 it would be way better than our current tolerance of +/- 0.1 for every single part that we are designing, no matter what size it is. I think if we put this in practice it will force the other designers to think a little more.

 

[CheckerHater]

fsincox,

Your phrase about “outside popularity” gave me idea to look into actual geographical popularity of certain standards. It took some time on the Internet (but not a lot of it though).

Here are the results: General tolerance standards around the world:

Brazil: NBR ISO 2768

England: BS EN 22768:1993 (ISO 2768:1989)

France: NF EN 22768:1993 (ISO 2768:1989)

Germany: DIN ISO 2768

Italy: UNI ISO 2768

Japan JIS B 0405 (ISO 2768)

Russia: GOST 308931-2002 (ISO 2768:1989)

Sweden: SS ISO 2768

Some trivia:

Brazil is known as the most “standard-loving” country. First country in the world to adapt new international standard for electrical outlets. I’ve heard story about sales person who had to run to local hardware store to buy new plug, otherwise he couldn’t do the presentation of his machine.

England - Not so long ago made important decision: abandoned Taylor’s principle A.K.A. “Rule 1” and fully embraced ISO “independence principle” framework.

France and Germany are responsible for most of EU’s GDP. And I mean MOST of it.

Italy - well, 2013 Dodge Dart is 2012 Alfa Romeo Giulietta.

Japan – don’t know where to start

Russia – not exactly technology leader, but it spreads over 1/7th of the total landmass - can’t ignore that.

Sweden – the largest European economy outside of Euro-zone.

So, there may be some popularity “outside”.

 

[pmarc]

JJAV1983,
If you are really confident that the system will work, use it.

 

[KENAT]

JJAV1983, as you've probably gathered the problem with wanting to use your tolerance block is that generally metric dimensions don't have trailing 0's. So you can't really do it unless you come up with your own exception to the standard or equivalent.

Having a default +-0.1 mm is indeed dubious unless all your parts are of a similar size and require similar precision.

Having a single value that you set for each part can work well and is what I did at my employer back in the UK in the early 2000's (never even heard of 2768).

As dgallup says any kind of 'default' tolerance system is open to abuse. Invoking a default without consciously considering what the default is and the implications of it can cause problems.

Typical American block where tolerance varies with number of decimals is also often abused, in which direction tends to depend on what the CAD default # decimals is set at.

Beyond that some can be taken in different ways for instance the way DiegoLGraves applies that clause in 2768 starts the undermine the idea of interchangeability which was the very reason for the original development of GD&T as I understand it.

Posting guidelines faq731-376

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

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