Position of an angled hole - Datum Reference Frame questions. 1

[Jacob Cheverie]

Please see the attached drawing that I quickly made. I am working on a component with similar geometry and I would like some feedback on a few questions.

1. The position of the Ø.750 hole references primary datum C and secondary datum B. This datum reference frame fully constrains all 6 degrees of freedom, correct? I believe that by adding a tertiary datum reference, say D, we would be adding no value and over defining the component.

2. Secondary datum pattern B is assigned a position with respect to datum D. It's typical (or should be) that there is a geometric tolerance on the pattern with respect to C, since the Ø.750 feature is relative to DRF [C|B|*]. Since there is no MMB/LMB modifier on datum reference B, is it OK to not assign a position on B with respect to higher precedence datum C? Would it be best to use multiple single segment position control to cover my bases?

3. What I've seen is one of the end points of the Ø.750 feature located with basic dimensions back to datum D. This would be invalid if we don't include datum reference D in the DRF of the Ø.750 feature, correct? The origin of DRF [C|B|*] will be in the plane of C, further constrained in rotation by the axes of the Ø.500 pattern, and located on.....

3.1. Which of the three Ø.500 features will I use to find the origin of DRF [C|B|*]?​

3.2. Should the basic dimensions locating the Ø.750 feature come from the origin of the DRF, which will be off the part? If so, is there a shorthand way of showing that on the drawing other than drawing the origin out in space?​

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1661267471/tips/SAMPLE_xau6hm.pdf[/url]

 

[jassco]

Hi, Jacob Cheverie:

The hole (dia. .750) is not positioned properly. It can "fly" out of plane of datum feature D.

I were you, I would reference datums as follows:
Datum feature D should be a primary.
Datum feature B (2 holes) can be secondary.
Datum feature C can be tertiary.

Or, alternatively, you can use datum as follows.

Datum feature D should be a primary.
Datum feature C can be secondary.
Datum feature B (2 holes) can be tertiary.

Best regards,

Alex

 

[Burunduk]

Jacob Cheverie,

1. Yes, when datums C primary and B secondary are simulated, all 6 degrees of freedom are fully constrained, and the tolerance zone for the Ø.750 hole is fully defined. But does this datum structure represent a functional requirement? Why reference these datum features and not others, and why in this order? Note that the primary datum feature for controlling a hole doesn't always have to be a face which is perpendicular to the hole and intersecting with it. It is only the case when a direct and tight control of the orientation of the hole relative to such face is required.

2. Although it often makes the most sense, you don't always have to follow the textbook method, in which the secondary datum feature is controlled with reference to the primary (and the tertiary with reference to the primary and secondary). But, some basic relationship and some geometric tolerncing link, even an indirect one, between datum features used in the same feature control frame has to exist. If for example, 80° is made basic, and an angularity tolerance is specified for datum feature C with ref. to datum feature D, then it is already legitimate (because the pattern of datum holes B is also controlled with ref. to datum feature D).

3. What you need is linking the controlled hole by basic dimensions to the datums the DRF is based on. You could use the intersection between the plane of datum feature C and the plane of datum feature D as a place from which a basic dimension is given to one of the holes in the pattern used as datum feature B. This will directly tie up the geometry of the 3 holes associated with datum B with datum plane C in a robust basic relationship and make the plane of datum D a legitimate origin for a height dimension to one of the end points of the Ø.750 feature. It wouldn't matter that "D" is not in the controlling DRF for the Ø.750 hole, you could think of it as a gage plane of sorts and forget about its association with D.

As a side note, a pattern of holes as an RMB datum is difficult to implement in practice, with equal priority between the holes in the pattern.

 

[jassco]

Hi,

If you look at SECTION A-A, this datum feature C has very little capability to control vertical position because of 80 degrees angle. In math, this is called "Ill-condition". Datum feature C is a very good candidate in controlling left and right, but it is horrible in controlling up and down (on SECTION A-A).

Best regards,

Alex

 

[Burunduk]

jassco,

I see where you are coming from, but the way I see it, it's not really the job of datum feature simulator C to constrain the "vertical" translation.
Let's talk about the 3 translational DOF;
I always imagine one of the 3 orthogonal planes of the DRF aligned with the primary datum. Since datum plane C is the primary, I imagine that it is related to a translation constraint, let's say X, in the direction perpendicular to it. Then datum feature holes pattern B and its simulator is related to the constraints of translation in axes Y and Z, the directions of which lie in the plane of datum C. Since the datum feature simulator for B is a set of 3 adjustable pins made to expand until the fullest possible contact with the actual holes, I wouldn't worry about their ability to constrain Y and Z. However if B was referenced at MMB, that would be completely different story and I would agree that it would result in a huge shift affecting the "height" of the inclined hole.

Now, with the above said, I know that there is often more than one way to establish the orientation and location of the DRF and its coordinate system based on a given set of datums, especially in non-orthogonal cases like these, and then one of the axes could point "upwards" (at 10° to datum C), BUT -this doesn't really change the nature of the constraints as I described above, which is more related to the geometry of the datum features and their true geometric counterparts, as well as their precedence order. The directions of the axes as chosen don't affect the end-result in terms of the amount of DOF constrained.

 

[3DDave]

What is 1/tan(10 degrees)? That's how much the hole will move relative to the base plate per variation of the holes relative to datum feature C or the variation in angularity between C and B. That's a factor of roughly 6:1.

X and Y may be in the plane of C but the datum feature simulators/TGCs absolutely aren't. Also, the first of the three holes to have it's "expanding pin" reach full contact will be the winner - the other two won't matter.

 

[Burunduk]

3DDave,
If the height of the ends of the inclined hole axis relative to the base plate is a primary concern, then the base plate should be the primary datum. That's why I asked the OP why the shown datum structure was chosen. Increased variation possible in specific directions relative to a face that is not part of the datum structure doesn't mean that any degrees of freedom remain unconstrained by the current scheme. And the amount of effective datum holes out of the pattern of 3 doesn't really matter.

 

[3DDave]

I was making the observation that jassco was right and your analysis was of the theoretical condition, not the as-manufactured part acceptance. You may not worry, but whoever makes these parts should.

 

[Burunduk]

3DDave,
You are conflating between two different issues. You are talking about the variation relative to a non-datum feature allowed by the given datum scheme, possible datum feature variations and the tolerance applied to the controlled hole. jassco raised a concern regarding the ability to constrain a translational degree of freedom by datum feature C, and that is what I addressed in my response to him. 

Both in theory and on an actual as-manufactured part, a primary planar-type datum feature and its simulator will be able to constrain the one translational DOF that it should be expected to constrain, and will not be able to constrain the two translational DOF that it shouldn't be expected to constrain (It is different with rotational DOF in case of rocking - but that's beside the point). As for the big picture concerning all 6 DOF, in any realistic scenario once you fully support the part on a planar datum feature simulator for datum feature C, and then expand pins inside datum holes B, the part will not be able to move up and down or in any other direction, and neither rotate. The only case in which this would be untrue is if the B holes and the C face are produced parallel or nearly parallel, which means that something is off by 10°  from the nominal orientation.

With that said, the datum structure does seem odd, and again, this is why asked the OP whether this suits function. Most likely not.

 

[3DDave]

I am not conflating anything. I understand what Jassco asked and you don't.

 

[Burunduk]

OK, sure.