Good Practice while stating tolerance on drawings!

[var10]

Hi Everyone,

We design and manufacture CNC milled parts. During the prototype stage we get them made up in China. Our manufacturer there uses DIN ISO 2768 Fine tolerances for milled metal parts. So I started to put them in our drawings under the tolerance area as - as per din iso 2768 f.

I am starting to think it is not a very good practice to ask the viewer of my drawing to find and open another document just to know tolerances. So I want to go back to the old way where it states xxdec it is +/- yy. Do you agree with me? or it doesn't really matter? Eager to hear your opinions. So either I have the entire din tolerance standard in my dwg or go back to earlier ways.

Thanks in advance!

V.

 

[powerhound]

While I am not really a fan of ISO, it is much better than locating features using +/- tolerances. Stick with GD&T. I prefer ASME but if ISO is what you have, then use it. So, to answer your question: no, I do not agree with you.

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

 

[cjccmc]

Asking that question on this forum = asking Allstate agent if you should have insurance

 

[MikeHalloran]

So I googled it, and found two handy tables, extracted by a couple of tool vendors, that give plus/minus tolerances.
If there's more than that to the spec, you need to provide a legit copy to your suppliers.

Personally, I'd rather the drawing stand alone, so the traveling documents in the shop don't have to include books, or worse, excerpts of books.



Mike Halloran
Pembroke Pines, FL, USA

 

[bxbzq]

Notorious iso 2768, I just hate it very much. Including the tables from iso-2768 in the drawings could be a solution. But many designers are not aware that the iso 2768 means much more than those tables.

 

[ajack1]

As the customer you can choose what standards you want to work to, the tolerances on the drawings, if you want to use imperial or metric, your payment terms, the list goes on.

From that companies will quote you a price if they want to work for you and you can choose to accept or reject it.

 

[CheckerHater]

I am starting to think it is not a very good practice to ask the viewer of my drawing to find and open another document just to know tolerances.

Interesting!
Do you specify Rule 1 on your drawings, or do you attach copy of ASME Y14.5 to every print?
How about ASME B4? Do you attach copy every time you place H7 or g10 on the drawing?
"The viewer of your drawing" is charging you arm and leg claiming the parts will be made to specifications. Let them learn specifications!

And BTW, it's not a big deal to make small tolerance chart part of your title block.

 

[KENAT]

There is nothing fundamentally wrong with referencing a separate spec - it gets done all the time for thread forms, surface roughness, GD&T... However, I do have some sympathy with what you're saying and especially if it's a spec that isn't readily available/the folks doing the work aren't familiar with. I also get Mike H's point and will admit for tolerancing at least I like the idea of seeing it on the drawing.

However, by the sounds of it you're just recording what the shop is doing anyway right?



A bigger issue is as the designer do you understand what tolerance you are invoking by referencing ISO 2768, and do you allow for them in your tolerance analysis?

Additionally have you read the entire spec and are you happy with some of what it says about not meeting it's tolerances not necessarily being cause for rejection?

The philosophy of ISO 2768 seems to be based on what a machine shop can readily achieve (or could how ever many years ago they did the study that the spec is based on), the philosophy of American drawing system (and historically the British system though as it has moved to ISO I'm not so sure any more) is more about what is required by end function.

(By the way, I believe it's been ISO 2768 for a long time, or are you deliberately referencing the old DIN?)

Posting guidelines faq731-376

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

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

 

[drawoh]

var10,

I am working to ASME Y14.5M-1994 in metric. Since we cannot use trailing zeros, the tolerance notes with the trailing zeros are not functional.

I tolerance each and every dimension, based on my requirements. I am aware of what my fabricators can do, but this is only to keep my from specifying the impossible. Tolerancing each and every dimension takes perhaps five more minutes per drawing, unless the tolerances do not work. In that, I have detected a serious design problem, and I can solve it.

--
JHG

 

[KENAT]

Yep, as drawoh suggests there are some advantages to actually assigning tolerances to each dimension/feature as you go through the drawing in that it makes you think about the required tolerance.

Over reliance on standard tolerances - be it the typical US block tolerance or something more like ISO2768 - without taking them into consideration wrt to fit & function etc. can cause all sorts of problems.

Posting guidelines faq731-376

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

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

 

[TheTick]

It's one thing to make a vendor look up rules or material specs. I would NEVER make a vendor look up tolerances in a separate document. I always make tolerances explicitly defined o he drawing.

I've been bitten before by similar things. Tolerances change in a separate document or template, and suddenly my parts don't fit because the tolerances changed.

 

[fsincox]

Tick,
I too have been surprised how this complicates things, particularly when you work to inch and the customer's part is really in metric and internal specs say I need to model all parts to the mean of the tolerance range!
Frank