A simple question on GD and T: Datum feature vs datum 1

[alexengineer]

Hi guys...
I have a simple question to verify my understanding of Datum feature and Datum.
Say.. a through hole on a solid block as datum feature. If we specify the surface of a hole as datum feature, am I right to say that that datum is actually the axis that run through the hole.


I am asking this question cos I have seen many drawings who people specify datum feature as the axis running through a hole. This to my humble understanding is conceptually wrong. Any experienced engineer who is willing to help me verify my understanding.

 

[desertfox]

Hi alexengineer

To specify the axis of the hole as a datum you must put the datum letter in the same plane and position as the dimension line but on the opposite side of the dimension line arrow or place the datum letter actually on the hole centre line itself.



|<--------->| datum letter here with box and triang
| | ular boxed arrow (directly opposite
arrowheads


regards desertfox

 

[TheTick]

If it is a straight cylindrical hole, the datum is the axis of the hole.

A common error is to specify the axis of a stepped bore or shaft. A datum can only be the axis of a single bore, and the designer must specify which surface is used to derive the axis.

Due to illness, the part of The Tick will be played by... The Tick.

http://www.EsoxRepublic.com

 

[MikeHalloran]

One small problem: There's no such thing as a straight cylindrical hole. At some level of precision, they're all not straight and not round.

From a functional standpoint, i.e., if you're actually trying to measure a part, the only datum that can be associated with a hole is the centerline of the largest gage pin that you can stuff into the hole without distorting the part or the pin.

The fundamental principle underlying GD&T is that a datum is a tangible thing, that you can touch in a reproducible manner. The axis of a hole is imaginary; there's nothing there to touch.

Yes, I've seen holes used as datums many times, and it usually resulted in a great deal of confusion at inspection time about where, exactly, the axis is. The answer is, it isn't anywhere.

-Mike-

 

[TheTick]

One small problem: There's no such thing as a straight cylindrical hole. At some level of precision, they're all not straight and not round.

Do tell! Here I spent all my money on perfect hole-makers. Maybe upon reading the entire post it became clear that straight was in contrast to stepped.

From a functional standpoint, i.e., if you're actually trying to measure a part, the only datum that can be associated with a hole is the centerline of the largest gage pin that you can stuff into the hole without distorting the part or the pin.

-OR- tapered floating pins used on sheet metal gages. Also, use of MMC on datum callouts allows use of MMC pins to locate cylindrical features.



Due to illness, the part of The Tick will be played by... The Tick.

http://www.EsoxRepublic.com

 

[tbuelna]

axleengineer:

As a rule of thumb, a datum should be some sort of tangible feature of the part (ie. not an axis of a hole). The GD&T on a drawing are there for dimensional inspection of the finished part, so the question you should ask yourself is: How do you locate something like a theoretical hole axis for inspection purposes? Without a CMM, it is quite difficult.

Back when I was in drafting class, they taught me that a drawing is a document for inspection purposes, not for manufacturing purposes. If you keep that in mind when you are making a drawing, it will help you to decide how to apply your GD&T appropriately.

You should also refer to AMSE Y14.5-2000, it gives a very good instruction on how to apply datums properly.

Regards,
Terry

 

[desertfox]

HI alexengineer

Have a look at this site

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

regards Desertfox

 

[alexengineer]

Wow.. Desertfox. YOur link is good...!!!
Thanks guys for this clarification.
This really helps clarify datum and datum features

 

[ewh]

Note the disclaimer on the above noted link.

"The information on this page has not been checked by an independent person. Use this information at your own risk."

Geometric tolerancing has been around in some form or other since the early part of the last century, so it's understandable that there are so many different methods and interpretations. The key is to agree on a STANDARD!

I can find nothing in ASME Y14.5M-1994 that corresponds to the "REFERENCE COMMON AXIS OR MEDIAN PLANE OF ALL FEATURES" example in the above link.

There are NO examples of a centerline as a datum in ASME Y14.5M-1994. ALL related examples are of an AXIS (or PLANE)of a DATUM feature.

So please, if anyone can show me where in ASME Y14.5M-1994 it is stated as an acceptable practice to label a centerline as a datum (not as an axis of a datum), let me know. I'll have many apologies to make.

Eric

 

[alexengineer]

"show me where in ASME Y14.5M-1994 it is stated as an acceptable practice to label a centerline as a datum (not as an axis of a datum)"
Quote

Eric... what u are saying is the terminology speaking, the centerline of datum feature is called axis of datum feature and not datum. I do agree with u on this...

The least that we agree on is that on 2D using GD and T, surface of a hole should be put with datum feature label and not the axis. It's probably terminology that u might probably be more accurate. But then again... datum existing in imaginery world for us to reference features to.. In this aspect, the axis of a datum feature could still be considered a datum(BUT definitely to be implied from datum feature and not to be stated as datum)

What do u think?

 

[MikeHalloran]

Terry's drafting instructor was correct, if perhaps not specific enough. The mists of time may have blurred the details, but the message stuck.

The purpose of a _sketch_ is to illustrate what you _want_.

The purpose of a drawing is to define what you will _not_accept_.

A good drawing will unambiguously, if indirectly, delineate how a part is to be set up for inspection, what measurements are to be made, and what acceptance criteria are to be applied to the measurements.

Y14.etc provides a powerful standardized shorthand for such delineation.

One thing I've noticed over the years; if people spend much time discussing what a given drawing's symbols really mean, it may indicate that that the person who did the tolerancing didn't fully understand how the part would be manufactured or applied, and just slapped on a layer of Y14.etc in the interest of posterior protection.

-Mike-

 

[ewh]

alexengineer,

We are in agreement. From a functional viewpoint, the axis is the element which is generally more important. If we are talking about a simple cylinder, there would be no confusion in calling the axis a datum. As soon as you have more than one diameter though, you need to know which diameter is being refered to, which is why a common axis should not be considered a datum. By not allowing an axis to be a datum, the standard eliminates this ambiguity.

Eric