Is this GD&T correct 1

[SiW979]

Hi People!

I'm looking at a drawing and am trying to convince myself that I'm reading it wrong, but I can't. It just doesn't make sense to me and everyone I've shown it to so far agrees, but we're not confident enough with our GD&T knowledge to say it's definitely wrong, as it'll be wrong on many drawings, and dates back nearly a quarter of a century, across a number of suppliers, none of which have said, "no, that's rubbish!".

Is it logical to specify a thickness of 3.9/3.8 (so, ± 50 microns) and have a flatness tolerance band on both sides of 250 microns. I suspect that the 250 microns should be 25 microns, but am not 100% confident. I'm thinking that based on the faxt that flatness is measured between two parallel planes, that the parallel frame is not required and I could just get away with the flatness tolerance? and that the drawing is over constrained? Or am I wrong? Also do I need a datum for a flatness tolerance?

Many thanks

Simon

Best regards

Simon NX7.5.4.4 MP8 - TC 8

http://www.jcb.com

 

[CheckerHater]

Your picture doesn’t show, most likely because you used “&” symbol in your file name.
Most important question is: are you working to ISO or ASME drafting standards?
In ASME world you are most likely right: flatness normally is a refinement of size tolerance.
Things are different in ISO world though.
You appear to be working for international company, so it would be better if you clarify the rules of your office.

 

[SiW979]

Hi we are working to ISO standards, I will try with the image again.

Cheers

Si

Best regards

Simon NX7.5.4.4 MP8 - TC 8 www.jcb.com

 

[CheckerHater]

we are working to ISO standards

In this case the drawing is correct.

ASME is using what is called “Envelope requirement” meaning that THE ENTIRE PART is supposed to fit within 3.9 envelope but LOCAL MEASUREMENTS, say using a caliper, should be no less than 3.8.

Under said requirement obviously your flatness cannot be larger than 0.10.

ISO operates under “Independence principle” which means, as you normally use two different tools to measure thickness and flatness, your thickness LOCAL MEASUREMENT with caliper should be within 3.9/3.8, and flatness measured with indicator INDEPENDENTLY from thickness should be no more than 0.25.

Judging from the appearance of your part (sheet metal-ish) one could imagine it to maintain its thickness well, but being slightly bent. So the ISO way will actually make sense. (If you are dealing with dowel being pressed into part, Envelope requirement actually works better).

This is why you can also specify Envelope requirement in ISO and Independence principle in ASME as necessary.

If my explanation wasn’t clear enough, many people on this forum will be glad to clarify the case to you.

 

[SiW979]

What about the parallelism of 0.08? is this feasible with the given flatness? is it even required?

Best regards

Simon NX7.5.4.4 MP8 - TC 8

http://www.jcb.com

 

[CheckerHater]

I am a bit confused about parallelism myself. Maybe they use it as refinement of thickness, controlling sides of the part being equidistant, but not necessarily flat or straight. I have to check with my books.

 

[PaulJackson]

Here is a way to specify in ASME 94 terms what "I think" the designer is trying to constrain.
Paul

 

[PaulJackson]

oops

 

[SiW979]

The way I see it, the dimensions for the thickness with the +/- 50 microns combined with the flatness tolerance, would cover the parralelism?

But I'm not that experienced so I could be wrong

Cheers

Si

Best regards

Simon NX7.5.4.4 MP8 - TC 8

http://www.jcb.com

 

[fsincox]

I have not seen that practice in years, I believe the ISO no longer supports it, although like many old ASME drawings practices they are often propigated and do not just magically disappear. I would love to see what people think it means, I once proposed to adopt it for our use in a company standard that is untill I read a book explaining how the ISO actually used it (I hope it was wrong).

 

[fsincox]

1) the flatness is required due to the lack of the envelope principle in ISO.

 

[fsincox]

no datum for flatness in either system!

 

[fsincox]

for flatness the "datum" is the concecpt of a perfectly flat plane (the concept of perfection is referenced, again)!

 

[pmarc]

Whatever the design intent is, the parallelism callout has been specifed incorrectly.

First of all, like in ASME, its value has to be greater than applied form tolerance. In this case it obviously does not full this requirement.

Secondly, parallelism tolerance requires a datum reference. Some can say that these two leaders indicate that one surface is datum feature for the other. And okay, that could make sense, if one of the leaders was terminated by triangle and not by arrow. Since it is not, it is pretty unclear which side serves as datum feature and which is toleranced one. The print could be much more specific about it.

And finally, such way of attaching feature control frame with both - datum feature and toleranced feature - was withdrawn by ISO 1101 in 2004 edition. I do not have a copy of this standard with me right now, but I am sure you can find proper statement in one of the appendices to the document.

 

[CheckerHater]

I agree with pmarc that parallelism is most likely wrong. It would make much more sense if its value was greater than flatness.

If the drawing itself was prepared before 2004, not using “triangle” termination is not that much of a sin.

Long ago it meant that choosing side for a datum is irrelevant. By calling one side of symmetrical part a “datum” you would imply that that side is somehow more important than other which might not be the case.
In fact, you could actually terminate the datum symbol with an arrow, as shown on the picture.

Naturally, all of this is very, very former practice.

 

[fsincox]

pmarc,
Sorry, but if it was done to a previous standard, if it in-fact references that previous standard is it still "wrong"? The ASME game is not played that way. Like I said this stuff doesn't magically disapear just because the standard changes. If "they" change the rules of the game, why are the people who tried to follow it to blame, that is very bad form, IMHO.

 

[SiW979]

Thanks for all your answers People

Much appreciated.

Si

Best regards

Simon NX7.5.4.4 MP8 - TC 8

http://www.jcb.com

 

[pmarc]

Frank,
Yes, if the print is dated before 2004, this way of tolerance frame presentation is acceptable (I mean, assuming that the triangle is there instead of one of the arrows).

 

[pmarc]

CH,
Does your latest example come from any official ISO document?
I have ISO 1101:1983 in front of me and there is something similar shown in Table 2 and in couple of other pictures, however only as an example of toleranced feature indication.
When it comes to datum feature identification there is always triangle required (filled or unfilled). Similar-to-op technique is shown for instance in figures 36 and 37. But the triangles are there.

http://files.engineering.com/getfil...-4808-bc6a-f0b8f0dd18c5&file=iso1101_1983.png

It is not that I nit-pick on arrows or triangles. I have been through cases when lack of proper datum feature identification, like here, created a lot of troubles.

 

[CheckerHater]

No, it’s not from official document. I don’t even have a year reference.
Like I said, it’s all things of the very distant past. Maybe it’s better that way.