Is Use of a Center plane for Tertiary Datum Legit? 3

[ringman]

This may be a bit difficult to describe in words alone, but here goes.

We have a rectangular plate, on which we have designated the surface as datum feature A, the bottom edge as datum feature B. So far no problem. The width has been designated as datum feature C which establishes the mid plane of the width. Now the tricky part. We/they have located 2 tooling holes with respect to primary datum feature A, secondary datum feature B, and tertiary datum feature C,(mid plane).

Any thoughts or comments on the legitimacy of this callout,as relates to the tertiary of datum C specifically?

 

[weavedreamer]

Sounds legit. Envisioning basic dimensions from the tooling hole c/l's to B and c/l of C with the holes perpendicularly normal to A.

 

[powerhound]

If I'm getting the right visual then what you have sounds okay, as long as the datum feature identifier is shown aligned with the dimension line. If the holes are shown centered about, or on, datum plane C then no dimension is required.

Powerhound, GDTP T-0419
Production Supervisor
Inventor 2008
Mastercam X2
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[drawoh]

ringman,

I take it you mean you applied a positional tolerance with respect to your three datums?

It sounds okay to me. Your datum C is a feature of size. If this is not an accurate feature, you are going to have to do some tricky tooling, or you are going to have to call it up at MMC. Is this acceptable?

JHG

 

[axym]

The callout is legal.

Just make sure that they are creating the datum reference frame correctly:

A: Flat plate (3 point contact)
B: Flat plate, at 90 degrees to the A plate (2 point contact)
C: Two adjustable parallel plates, at 90 degrees to both the A and B plates (generally 1 point contact on each plate)

Regarding legitimacy, what was the reasoning for the datum feature selection? Why was it one side for B and both sides for C? One concern is that contracting the vise-like simulator for C might pull the part off of its proper 2-point contact with the B plate. The likelihood of this depends on the relative lengths of the B and C sides, and their relative squareness.


Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[TheTick]

Legit, yes. Likely to be confucing to the underschooled. Be prepared to do much 'splaining.

 

[fcsuper]

Let me parrot everyone else. Yes, it's good. However, if you do have concerns about it, I would suggest it is acceptable to revisit the scheme to see if alternative methods for datums exist.

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

 

[drawoh]

ringman,

I am trying to visualize your drawing.

Datum[ ]C should be applied to the width dimension. Once you have done that, you can draw a centre line up the middle of your plate, and apply basic dimensions to it. Forget about [±][ ]dimensions. Consider doing this all some other way if the width dimension is not accurate.

JHG

 

[CheckerRon]

Sounds acceptable to me too, except I have had similar parts where I would be using the plate width as datum B (secondary) and the bottom edge as the tertiary.

So Datum B is my symmetry datum and tooling holes would be in-line with the center of datum B. That way datum B can be clamped 2 points on each side for stability and the machine and/or CMM can find the center axis.

If the tooling holes were dead center, they would make a great datum B pattern datum.

 

[tost]

Ringman,
I am not able to visualise Datum B since you have defined it as bottom edge, what this means are you trying to define an edge as Datum ?

 

[KENAT]

I agree that fundamentally it sounds legit per ASME Y14.5M-1994, 5.4 is an example of using pos tol WRT Datum Feature Center Planes.

Is the attatched something like what you have?

I've also shown my understanding of what drawoh says about the datum feature at MMC.



KENAT, probably the least qualified checker you'll ever meet...

 

[drawoh]

KENAT,

Your drawing with the alternate FCF is exactly what I am visualizing.

Let's see what ringman says.

JHG

 

[CheckerRon]

What Ken has drawn is exactly what I was talking about also in my post above.

The only difference being that I like to see primary, secondary & tertiary datums in alphabetical order (e.g. A,B,C), though it doesn't matter what character is used.

 

[ringman]

Ringman says;

That is close on to what I had attempted to describe.
Perhaps a small point, but my part is a LOT bigger.

My concern is that the mid-plane of datum feature C is established TOTALLY independent of the other 2. Consequently, in theory will not be perfectly perpendicular to datum feature B.

So whens it comes time to establish the datum reference frame A, B, C, we are trying to center to a line, rather than a point and is essentially non-achievable.

I have had discussions with some who believed he mid-plane of C is influenced, or driven by B, and that is a point of contention in the discussion,
or has been.

 

[KENAT]

Ron, my print is in alphabetical order, or am I missing something? Like you I've seen the 'width' datum used as the secondary more than I've seen it as tertiary, however tertiary is what ringman says.

ringman, I think swapping the order of your B & C as Checker Ron mentions will solve the problem. In this case what is currently B will be single point contact so the orientation issue is taken care of by the 'width' datum.

See amended sketch.

KENAT, probably the least qualified checker you'll ever meet...

 

[KENAT]

Oh yeah, as to your size issue, is it really worth even mentioning? I suppose it might make a difference to inspection methodology but that should't be a major factor in the dimensioning scheme, should it?

KENAT, probably the least qualified checker you'll ever meet...

 

[drawoh]

ringman,

I suggest that you be careful about discussing imaginary features. There is no centre plane on your part.

Let's use KENAT's drawing as a reference.

Your datum[ ]A, the bottom face, locates your part in one linear plane, and two rotation planes. Your datum[ ]B, the lower edge, locates the part in one linear plane, and one rotation plane.

Datum[ ]C, your width, locates your part in one linear plane, only. It is the tertiary datum. Consider the possibility that your width is grossly non-perpendicular to the lower edge. A fixture might pick up the top corner on one side of the width, and the lower corner on the opposite side.

If the centre defined by the width matters to you, you must follow CheckerRon's suggestion, and make the width your secondary datum, datum[ ]B. A fixture will pick this up at two points on one side, and one point on the opposite side. Datum[ ]B will be perpendicular to datum[ ]A. Your lower edge, datum[ ]C, is picked up at one point, somewhere.

If the centre really, really, really matters to you, make it your primary datum. On a plate, this will be very difficult to inspect.

As I noted above, your whole model is a mess unless your width is accurate. Using your two tooling holes as datums[ ]B and[ ]C is a good idea. Use profile tolerances to control your outline.

JHG

 

[ringman]

The alpha sequencing of the datum features is dependent strictly of the design requirements and not alpha sequence driven.

The sketch that Kenat provided, clearly illustrates C as the width of the part. Therefore the related datum would be the mid or center plane of that feature independent of the other datum features.

I would agree that if we changed the sequence of the datum features to ACB we would have a legitimate callout, however that is not in compliance with the situation of the drawing I am concerned with.

In the situation I am concerned with datum feature is the orienting feature and C is used for centering of the holes.
Additionally C is RFS.

 

[ringman]

Drawoh

I believe that datum features are real and not imaginary.
I forgot to add that in the previous posting.

 

[KENAT]

By the way I don't see that C (width tertiary per my first sketch) is totally independant. I think drawoh explains it well above, I've attatched my own effort based loosely on fig 4-13 of the standard (some annotation left out).

KENAT, probably the least qualified checker you'll ever meet...