General Tolerances

[vctech]

I have been trying to find general angle tolerances to put into drawing title blocks but cant find any. What are general tolerances for angular tolerances of machined and bent angles? Can anyone help with what a standardized value for these would be?

 

[KENAT]

So, lets get the painfull part out of the way.

TOLERANCES SHOULD BE DECIDED BASED ON FUNCTION FIRST.

Sorry for shouting but questions like this will bring up that old debate. Most posters here subscribe to that philosophy and many feel that 'block tolerances' or 'standard tolerances' or the like are over relied upon by many designers/engineers to the detriment of performance etc. There are plenty of threads bemoaning that if you want to look.

So, please explain what you mean by General tolerances.

Are you more interested in typical process capabilities? Again quite a few threads on this and no simple answer as it varies so mucy by vendor/machine capability, part size, part material etc.

If you just want examples of what peoples block tolerances are my place says +-0.5° which I think is fairly typical.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[KENAT]

thread1103-252017, and shy away from even thinking of using ISO2768, at least in my not so humble opinion.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[SteveMartin]

I agree with Kenat. I'm slowly trying to turn my engineering group away from reliance on block tolerances.

For what it's worth my current company's default angle tolerance is +/- 1 degree, and I've seen as tight as 0.5 and as loose as +/-5 (both in degrees), although I had a very hard time believing the +/-5 would ever produce consistently acceptable parts for any application.

 

[drawoh]

vctinc,

What do you mean by "General Tolerances"? I have title blocks set up to say ±1° as the default. I could have just as easily set it to ±0.5°.

The ANSI/ASME Y14.5 standard, at least since 1982, has not allowed trailing decimals on millimeter dimensions. This means that you must set tolerances or geometric controls on each and every metric dimension. This is not particularly difficult, and as noted above, it leads to better quality drawings.

JHG

 

[juergenwt]

VC- ISO/DIN 2768-1 here:

http://www.ams-prototipi.it/webfiles/iso.pdf

 

[MintJulep]

A few more threads on the topic:

thread1103-222161

thread1103-199111

 

[KENAT]

I was hoping no one else would bring it up but they did so...

Be very cautious with ISO 2768. There as several threads, for example thread182-247563, discussing/questioning this thread for several reasons that can perhaps be summarized as:

1. The information used to create the tolerance classes etc. is probably out of date.

2. Conceptually it's basically saying that manufacturability is more important that function.

3. There is confusion on how aspects of it are applied/interpreted.

4. There is a get out clause at the end of the standard that says just 'cause a part doesn't meet the tolerance stated doesn't mean it should be rejected but you should look at function. Some believe this is just to allow for production permits/deviations whatever you call them. However it kind of conflicts with the idea of the drawing being driven by function. If the drawing doesn't record functional requirements what does?




KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[fcsuper]

In inches, common tolerances employed for machined parts is .XXX = +/-.005 and .XX = +\-.01 with angles either being 1deg or .5deg. .005 is roughly a .125mm (or just .13mm) and .01 is roughly .25mm. I've seen metric general tols like this, and they are technically "allowed" but not supported by an via ASME Y14.100's paragraph 1.2 and not specific statements regarding general tolerances which are found in one of the early paragraphs in ASME Y14.5. If you wish to use general tols with metric, you'll need to fully document that in your company's drafting standard, or at the very least, be very clear on how general tols apply on your drawing title block.

As mentioned, general tols should be used with caution, and there is no one size fits all.

For example, with sheet metal, you cannot really use a general tol unless you have flat parts or long flat areas. The reason is that very bend, cut or other feature is affected by the feature it is being dimensioned from. One bend may put your tolerance at +/-.016, but dimensioning across two bends immediately doubles that.

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

 

[MintJulep]

Please don't say "tolerance" when you mean "expected distribution of process variation".

 

[dgallup]

QUOTE
drawoh (Mechanical)
11 Sep 09 15:35
vctinc,

What do you mean by "General Tolerances"? I have title blocks set up to say ±1° as the default. I could have just as easily set it to ±0.5°.

The ANSI/ASME Y14.5 standard, at least since 1982, has not allowed trailing decimals on millimeter dimensions. This means that you must set tolerances or geometric controls on each and every metric dimension. This is not particularly difficult, and as noted above, it leads to better quality drawings.
------------------------------------------------------------
My company designs everything in metric & we have default tolerance blocks on every drawing. We just use a single value for linear dimensions & one for angular dimensions. Problem solved. I hate the ISO default tolerance blocks where if your part changes from one size group to the next the tolerances jump up or down.

And I frequently use +/-5 degrees for angles, got tired of the QC inspectors rejecting good parts because the 45 degree x .5 max chamfer was 48 degrees.

I like relatively loose default tolerances, if a dimension needs to be tight then call out what you need. And if it's really important use a special characteristic symbol to call attention to it. Way to many lazy drafters making impossible drawings with tight default tolerances and every dimension nominal with no other tolerance.

 

[juergenwt]

As long as you produce in the US only and your customers are OK with that - no problem.
If you are manufacturing to ISO spec's and sell to the rest of the world or for many international US companies -big problem.
You would not want to use the same tolerance field for a 60mm dia that you are using for a 12mm dia , would you?
If that is what you are doing - good luck to you.

 

[KENAT]

juergenwt the rest of the world doesn't strictly adhere to ISO2768 only. I can think of at least one place where it never even came up.

Posting guidelines faq731-376

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

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

 

[fcsuper]

Yeah, I think U.S. engineers have the false impression that "the world" is 100% metric in every thing. Given some of the pro-metric arguments I've heard in the past, you'd think that Europeans are born with millimeters imprinted on their foreheads, and every eyelash is exactly 1cm long.

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

 

[KENAT]

I don't recall ISO2768 or the associated concepts ever really coming up back where I worked in the UK. We worked to BS 308 then BS8888. Looking at the list of contents of BS888 (essentially a compendium of relevant ISO standards) I don't even see 2768 unless Im missing something.

http://www.roymech.co.uk/Useful_Tables/Drawing/Drawing.html

Posting guidelines faq731-376

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

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

 

[dgallup]

juergenwt - I have worked for American, British & French firms with customers all around the world. Never had a problem with this scheme, in fact it originated in England with a multi-national company. To me, it is nonsensical for the default tolerance to jump just because a dimension grew from 29.5 mm to 30.5 mm. The French in particular thumbed their noses at ISO anything, they use their own except for pipe threads, then they use good old NPT.

 

[juergenwt]

dgallup - If it works for you - fine. Nobody is forced to work to ISO.
It is just that I would rather work to a worldwide standard than work with something used by a select number of manufacturers.

 

[KENAT]

I think that's the point we're making. It may be 'approved' or what ever the term is internationally, but it's far from implemented.

ISO2768 is based on a DIN standard, and while I haven't seen a massive sampling, I've only seen it used on German drawings that I can recall.

Posting guidelines faq731-376

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

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

 

[Bishbosh]

BS8888 does call up BS EN 22768 which it states is identical to ISO 2768 and replaces BS4500.
I work for a major UK Defence company, and by referring to BS8888 on the drawing, I must work to the relevant Tolerance Standard.

 

[KENAT]

Well thanks Bishbosh, I didn't see it when I was looking at the list. Still don't see it but guess I missed it or the lists I can find are incomplete.

Interesting, I worked UK defence too and the tolerance concepts of iso2768 or the BS EN equivalent never came up, either in our drawings or any of our customers/suppliers. This was 99-04, I looked through BS8888 on numerous occasions and don't recall seeing it.

I still don't like 2768 though;-).

Posting guidelines faq731-376

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

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