position and basic dimensions 2

[joebk]

I think I know the answer to this but I am not 100% sure. Here is a description of the drawing I am checking;

Imagine a flat circular plate of steel. datum A is one surface of the part. The secondary and tertiary datums are defined by a central hole (datum B) and another hole on the part (datum C).

We weld a rectangular bracket onto the part. A reference dimension for the width of the bracket is included on the drawing and positional tolerance (no modifiers) W.R.T. A,B,and C is called out inline with the width dimension. This controls the location of the bracket in the "X" direction.

The sticky part is that there is a dimension with a tolerance relating to another hole (not a datum feature) on the part controlling the location of the bracket in the "Y" direction.

The question is - with the positional tolerance, is it valid to have a dimension and tolerance for the bracket as opposed to a basic dimension?

I think our drawing is valid because of the location of positional tolerance (inline with width dimension of the welded bracket). If I understand correctly this indicates that the position of the central plane of the rectangular bracket is controlled W.R.T. the defined datums. So this can only control position in one ordinate direction.

Of course I do not claim to be a GD&T Pro and this wouldn't be the first time I was totally wrong!

Thank you in advance for your help.

JBK

 

[Heckler]

We weld a rectangular bracket onto the part. A reference dimension for the width of the bracket is included on the drawing and positional tolerance (no modifiers) W.R.T. A,B,and C is called out inline with the width dimension. This controls the location of the bracket in the "X" direction.

The question is - with the positional tolerance, is it valid to have a dimension and tolerance for the bracket as opposed to a basic dimension?

If you're using a positional feature control frame to datums A, B & C for the location of the rectangular bracket then the dimensions that located must come from those datums. If you're only controlling in one direction you may need to remove a datum from the FCF positional tolerance. It would help if you posted a picture of the drawing.

Heckler
Sr. Mechanical Engineer
SWx 2007 SP 3.0 & Pro/E 2001
XP Pro SP2.0 P4 3.6 GHz, 1GB RAM
NVIDIA Quadro FX 1400
o
_`\(,_
(_)/ (_)

(In reference to David Beckham) "He can't kick with his left foot, he can't tackle, he can't head the ball and he doesn't score many goals. Apart from that, he's all right." -- George Best

 

[dingy2]

Since you are controlling the rectangular bracket in only 1 direction, use positional tolerances on the width or length of the feature. The dimensions to this feature on this axis would be basic. It would indicate that we are interested in that X axis only and would control the location of the feature on the width, its angle and perpendicularity. The feature would be confirmed using CMM and not checking fixtures.

The Y axis is not covered by the feature control frame so the dimension from another hole would have to have a tolerance and not be in a basic dimension.

Hope this helps.

Dave D.

http://www.qmsi.ca

 

[joebk]

Today is somewhat slow so I blew a couple minutes putting the following site together. I hope the link works, the page should explain the situation.

link

Note that this is not really the drawing but a close representation of it. I know that there is A LOT of information missing in the drawing and probably some errors.

 

[dingy2]

It is exactly what I have stated but the feature should have a width dimension with tolerances and not as a reference.

Dave D.

http://www.qmsi.ca

 

[ewh]

The width dimension may remain a reference dimension if it is specified elsewhere as a toleranced dimension (such as in a separate detail).

 

[Heckler]

You need to dimension from datum B and remove datum C from the Feature Control Frame. Datum C seems redundant for this simple of a part.

Heckler
Sr. Mechanical Engineer
SWx 2007 SP 3.0 & Pro/E 2001
XP Pro SP2.0 P4 3.6 GHz, 1GB RAM
NVIDIA Quadro FX 1400
o
_`\(,_
(_)/ (_)

(In reference to David Beckham) "He can't kick with his left foot, he can't tackle, he can't head the ball and he doesn't score many goals. Apart from that, he's all right." -- George Best

 

[dingy2]

Heckler:

Don't think so. That is removing datum C. We need it for orientation.

Dave D.

http://www.qmsi.ca

 

[Heckler]

Dave - Look at

Section 4.4.2 parts with Cylindrical datum Features.

fig 4-5a-b.

Section 4.4.2.1 Cyclindrical Datum Features


Heckler
Sr. Mechanical Engineer
SWx 2007 SP 3.0 & Pro/E 2001
XP Pro SP2.0 P4 3.6 GHz, 1GB RAM
NVIDIA Quadro FX 1400
o
_`\(,_
(_)/ (_)

(In reference to David Beckham) "He can't kick with his left foot, he can't tackle, he can't head the ball and he doesn't score many goals. Apart from that, he's all right." -- George Best

 

[PaulJackson]

The center-plane of the width of the bracket (2.042) must lie between two parallel planes spaced .010 apart from a datum reference frame established by A|B|C.

If there is missing information as noted <N1> "MAJORITY OF DRAWING AND CORPORATE INFORMATION REMOVED" then those that insist that reference dimensions refer to toleranced dimensions on the same level drawing can assume that the toleranced value exists but is not shown and those that are used to modern product drawings resembling former "process sheets" can assume that all the necessary information to interpret (decode) the callout has been shown.

I'm an old guy I'll let you guess which camp i'm in.

Paul

 

[joebk]

I think I have to have datum C because the bracket needs to be oriented with respect the the 1/2" hole (section 4.4.3 covers this). Otherwise the bracket can "rotate" around datum axis B and still be in spec. Check out the first sentence in section 4.4.2.3.

If I only use the center hole as a datum it does create the secondary and tertiary datum planes. But as I understand it, the secondary and tertiary datum planes can rotate about the datum axis so the features can float around the axis as well.

We use positional callouts similar to this (only 2 datums) all the time for bolt circles that are evenly spaced about the axis and do not require orientation with a pilot diameter.

The bracket is fully detailed in the drawing of the bracket itself so the width is specified as a reference only dimension in this case. Since there are no modifiers in the FCF I didn't figure this would be a problem. I don't really want to get into all the nasty details of the bracket itself.

After some research I don't think the Y dimension needs to be basic. Examples and info on pages 141-143 of the current standard seem to support this.

For the Y dimension to be basic I think I would need to add another positional callout on the thickness dimension of the bracket.

Conversely I think if the positional tolerance was called out with a leader to the bracket (and not a size dimension) it would control the position in X and Y and require basic dimensions for each.

Hopefully my interpretation of the standard is not to far off base!

Sorry for the long post and thanks to all for the help!

 

[dingy2]

Heckler:

Yes you may be correct about only needing datum B hole as per fig. 4.5 in the standard. The other hole is assumed to be on the same axis as the rectangular feature.

Certainly, if the hole was off the axis, one would need a tertiary and a basic dimension to the centre line would be required.

Would a tertiary datum hurt in this case? No, it wouldn't and there is no conflict so would I might still use a tertiary in this case.

Dave D.

http://www.qmsi.ca

 

[PaulJackson]

With the information provided I disagree that datum C need not be referenced to tolerance the rotational position of the bracket to C. I agree with Joebk.

There is an exception which would control the rotational alignment of the two...but it would be controlled indirectly via rules governing simultaneous requirements.

For that to be valid however, both of the feature (C and the bracket width) would have to have position tolerance callouts with identical datum and modifier references.

I disagree with the extension of this principle that constrains "degrees-of-freedom" that are not otherwise constrained by the common reference. It in effect makes all commonly toleranced features "one pattern". The rule however, is in the standard so it doesn't matter whether I disagree with it. I suspect the rule was originally intended to make "datum shift" allowances common to all identically referenced features but it got corrupted by those pondering "rocking plane" datum features so its definition was extended to constrain all six degrees of freedom among the "one pattern".

Paul

 

[ringman]

joebk,

Is it possible for you to post a link to the part that will interface with this that will show us the relationship required for the holes. I submit that what you have indicated in the link is slightly 'mis-directed' dimensionally, but would have to see mating part to know for sure.

Is the center hole or the pair of outermost holes most critical at next assembly?

Thanks
ringman