I have seen a number of drawings where a hole ends up with a larger tolerance east-west than north-south, so rather than expand the diametrical tolerance to cover both, I see two reference linear dimensions with separate control frames. (See top example) I was wondering if possible to combine these into two single segment control feature controlf frames on the diameter dimension. (See bottom example). Is this allowed and is it clear?
Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
2 Single Segment FCF to make Rectangular Tolerance Zone? 3
- Thread starter JasonSp
- Start date
Tks.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Prof,
A review of one of the leading GD&T texts shows the datum identifier used in conjunction with the targets. It seems to clarify the precise intent of the targets.
In the illustration it is not clear that the points X are used as a width. It would be clarified if the triangle (datum identifier)were associated with X1 and X2 as a width.
A review of one of the leading GD&T texts shows the datum identifier used in conjunction with the targets. It seems to clarify the precise intent of the targets.
In the illustration it is not clear that the points X are used as a width. It would be clarified if the triangle (datum identifier)were associated with X1 and X2 as a width.
Ringster, what else could the opposed points be taken as other than a centering datum? In the texts you refer to, does it show the datum identifier symbol attached to the drawing centerline? That tends to creat even more confusion as drawing users may think it's the centerplane based on the width of the opposed faces rather than the opposed points. Could you post a pic so that I can see your proposal? Tks.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
On the first page of ProfDon's casting drawing, it says "datum targets were used to establish datum planes Z, Y and X". What exactly is datum plane X?
Does this mean that the datum extracted from the two X datum target points is a plane?
Evan Janeshewski
Axymetrix Quality Engineering Inc.
Does this mean that the datum extracted from the two X datum target points is a plane?
Evan Janeshewski
Axymetrix Quality Engineering Inc.
Given the DRF Z/Y/X, Z establishes the primary datum plane which arrests 3 dof (a translation & 2 rotations), and Y establishes the secondary datum plane perpendicular to the first datum plane and eliminates one translation and the final rotation. All that's left is for datum X to constrain that last dof. Given that all dof are eliminated, you have a fully constrained part; a fully constrained part has three mutually perpendicular datum planes. What else could datum X be other than a plane? Consider that the two datum target points establish a point in space which is centered between the two datum target points and is simultaneously located on the two preceding datum planes. That center point thereby establishing the location of the third datum plane which is oriented with mutual perpendicularity to the two preceding planes.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
powerhound
Mechanical
I see the DRF in ProfDon's attachment as the primary datum establish by the 3 datum areas identified by the 3 Z's. The secondary datum is a datum line identified by single Y1 and being that it is a datum line, it restricts 2 degrees of freedom. The tertiary datum is the plane that runs between X1 and X2, perpendicular to Y1. Do I understand this correctly?
Powerhound, GDTP T-0419
Production Manager
Inventor 2009
Mastercam X3
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II
Powerhound, GDTP T-0419
Production Manager
Inventor 2009
Mastercam X3
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II
Yes, Powerhound.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Jim,
You stated that a fully constrained part has three mutually perpendicular datum planes. Does that mean that a part that is not fully constrained does not have three mutually perpendicular datum planes?
The two datum target points establish a center point that establishes the location of the third datum plane. So is the tertiary datum a point or is it a plane?
Evan Janeshewski
Axymetrix Quality Engineering Inc.
You stated that a fully constrained part has three mutually perpendicular datum planes. Does that mean that a part that is not fully constrained does not have three mutually perpendicular datum planes?
The two datum target points establish a center point that establishes the location of the third datum plane. So is the tertiary datum a point or is it a plane?
Evan Janeshewski
Axymetrix Quality Engineering Inc.
Evan,
There has been considerable debate for a long time about your first question.
Per '94, 4.1; "... These elements exist within a framework of three mutually perpendicular intersecting planes known as the datum reference frame." That does not distinguish between fully constrained & partially constrained, so therefore the argument is that once a single datum is referenced, three mutually perpendicular planes are established. To me, that implies that everything is set in stone (so-to-speak) with a single datum (point, axis or line). More correctly, only a discrete number of those three planes has been located in space (one plane for a planar datum ; two planes for a datum axis; three planes for a point datum, with a conical datum feature establishing an axis and a plane normal to it), and therefore only those located planes can be used as an origin of measurement. The spatial location and orientation relationship of each datum plane is established by the DRF. It's late, so I hope I've answered that somewhat coherently.
For your second question, the two opposed datum target points will establish a datum point which has its own three mutually perpendicular datum planes, one of which establishes the location of the third datum plane of the DRF, thereby eliminating the final dof.
A lot of this is tantamount to the chicken & the egg. A datum point establishes three mutually perpendicular datum planes intersecting at that locale ... or does the intersection of the three mutually perpendicular datum planes establish a datum point? A datum axis establishes two mutually perpendicular datum planes which intersect at the axis ... or does the intersection of two datum planes establish a datum axis? Per 4.4.2, "A cylindrical datum feature is always associated with two theoretical planes intersecting at right angles on the datum axis. ... This axis serves as the origin of measurement from which other features of the part are located." So which came first, the two planes or the axis ... is totally irrelevant beyond a theoretical debate. We measure from the datum simulators.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
There has been considerable debate for a long time about your first question.
Per '94, 4.1; "... These elements exist within a framework of three mutually perpendicular intersecting planes known as the datum reference frame." That does not distinguish between fully constrained & partially constrained, so therefore the argument is that once a single datum is referenced, three mutually perpendicular planes are established. To me, that implies that everything is set in stone (so-to-speak) with a single datum (point, axis or line). More correctly, only a discrete number of those three planes has been located in space (one plane for a planar datum ; two planes for a datum axis; three planes for a point datum, with a conical datum feature establishing an axis and a plane normal to it), and therefore only those located planes can be used as an origin of measurement. The spatial location and orientation relationship of each datum plane is established by the DRF. It's late, so I hope I've answered that somewhat coherently.
For your second question, the two opposed datum target points will establish a datum point which has its own three mutually perpendicular datum planes, one of which establishes the location of the third datum plane of the DRF, thereby eliminating the final dof.
A lot of this is tantamount to the chicken & the egg. A datum point establishes three mutually perpendicular datum planes intersecting at that locale ... or does the intersection of the three mutually perpendicular datum planes establish a datum point? A datum axis establishes two mutually perpendicular datum planes which intersect at the axis ... or does the intersection of two datum planes establish a datum axis? Per 4.4.2, "A cylindrical datum feature is always associated with two theoretical planes intersecting at right angles on the datum axis. ... This axis serves as the origin of measurement from which other features of the part are located." So which came first, the two planes or the axis ... is totally irrelevant beyond a theoretical debate. We measure from the datum simulators.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Jim,
I appreciate the input. It's always interesting to get another practitioner's views on fundamental questions that I'm not sure about. I credit (blame?) DesignBiz for this - the recent thread with the skewed A-B cylinders has made me question my whole belief system regarding datums.
So you're basically saying that if a FCF has at least one datum feature reference then it has a DRF. I suppose that makes sense.
When you state that "the two opposed datum target points will establish a datum point which has its own three mutually perpendicular planes, one of which establishes the location of the third datum plane", that's where I get lost. I'm not saying it's wrong, but I'm not able to reproduce the logic of it.
Your comments about the chicken/egg nature of datum points and axes with associated theoretical planes were interesting. If you can make sense of the theoretical planes, you're a better man than me. I can see it for primary datum features, but for secondary and tertiary I just can't.
You're final comment that "we measure from the datum simulators" made the most sense to me.
Evan Janeshewski
Axymetrix Quality Engineering Inc.
I appreciate the input. It's always interesting to get another practitioner's views on fundamental questions that I'm not sure about. I credit (blame?) DesignBiz for this - the recent thread with the skewed A-B cylinders has made me question my whole belief system regarding datums.
So you're basically saying that if a FCF has at least one datum feature reference then it has a DRF. I suppose that makes sense.
When you state that "the two opposed datum target points will establish a datum point which has its own three mutually perpendicular planes, one of which establishes the location of the third datum plane", that's where I get lost. I'm not saying it's wrong, but I'm not able to reproduce the logic of it.
Your comments about the chicken/egg nature of datum points and axes with associated theoretical planes were interesting. If you can make sense of the theoretical planes, you're a better man than me. I can see it for primary datum features, but for secondary and tertiary I just can't.
You're final comment that "we measure from the datum simulators" made the most sense to me.
Evan Janeshewski
Axymetrix Quality Engineering Inc.
Evan,
Re the two datum target pts establishing a datum pt, try this mental exercise;
First, poing your two index fingers at each other, as if using them as a set of opposed datum target points. Next, visualize a line between the tangent points (i.e. closest points to each other). Accepting that opposed features establish a center value (plane, line, point) of some kind is the first step; this is no different from working with FOS (Features of Size). So, at the center location between these two datum points, you have one of the following; a plane, an axis, or a point. A plane requires the establishment of three points in space (minimum), so the two opposed points don't establish a plane directly. Similarly, a line (axis) requires two points in space; again, not what you have at that center value. All that is left is a single point as the center value. That datum point has three mutually perpendicular planes associated with it, all meeting at the datum point. The problem most of us face is what we do with the datum planes that we're not using from each of the datums. In this example, Datum Z establishes one plane, Datum Y establishes one plane, but Datum X establishes three planes while only one is needed to complete the DRF. So we use the one which is mutually perpendicular to Datums Z & Y as the third origin of measurement, and ignore the other two.
Here's something similar, though not quite the same.
Datum A establishes the first datum plane, Datum B gives you an axis with its two datum planes ... that's three datum planes and three origins of measurement already. Datum C, however, adds a fourth datum plane ... which is not used as an origin of measurement, but only to lock down the orientation of the planes established by Datum B. I included this graphic because it is often overlooked that Datum C is indeed a datum plane, but it happens (in this case) to be coplanar with one of the Datum B planes.
Hopefully that helps rather than muddles things.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Re the two datum target pts establishing a datum pt, try this mental exercise;
First, poing your two index fingers at each other, as if using them as a set of opposed datum target points. Next, visualize a line between the tangent points (i.e. closest points to each other). Accepting that opposed features establish a center value (plane, line, point) of some kind is the first step; this is no different from working with FOS (Features of Size). So, at the center location between these two datum points, you have one of the following; a plane, an axis, or a point. A plane requires the establishment of three points in space (minimum), so the two opposed points don't establish a plane directly. Similarly, a line (axis) requires two points in space; again, not what you have at that center value. All that is left is a single point as the center value. That datum point has three mutually perpendicular planes associated with it, all meeting at the datum point. The problem most of us face is what we do with the datum planes that we're not using from each of the datums. In this example, Datum Z establishes one plane, Datum Y establishes one plane, but Datum X establishes three planes while only one is needed to complete the DRF. So we use the one which is mutually perpendicular to Datums Z & Y as the third origin of measurement, and ignore the other two.
Here's something similar, though not quite the same.
Datum A establishes the first datum plane, Datum B gives you an axis with its two datum planes ... that's three datum planes and three origins of measurement already. Datum C, however, adds a fourth datum plane ... which is not used as an origin of measurement, but only to lock down the orientation of the planes established by Datum B. I included this graphic because it is often overlooked that Datum C is indeed a datum plane, but it happens (in this case) to be coplanar with one of the Datum B planes.
Hopefully that helps rather than muddles things.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Ringster,
I don't have GeoMetrics III. Pls post a quick hand-sketch that I can look at. tks.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
I don't have GeoMetrics III. Pls post a quick hand-sketch that I can look at. tks.
Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
Jim,
I can see how the center point is extracted from the two datum target points. The three mutually perpendicular planes associated with the center point still didn't make sense to me though. Part of the problem for me is that the "center point between two datum target points" does not have the same constraint properties as a center point extracted from a spherical datum feature.
One way that I evaluate my understanding of an issue is to imagine teaching it and dealing with the questions that would arise. Every class seems to have one or two students who just have to know where everything comes from and won't just take the instructor's word as gospel. If one of those students asked me to explain the logic of imagining three theoretical planes and then ignoring two, I couldn't.
So I read through the DRF chapter in one of the public review drafts of the new standard. The description of cylindrical datum features has significant changes.
1994: A cylindrical datum feature is always associated with two theoretical planes intersecting at right angles on the datum axis.
2009: A primary cylindrical datum feature is always associated with two theoretical planes intersecting at right angles on the datum axis. Depending on planes established by higher precedence datums, secondary and tertiary datum axes may establish zero, one or two theoretical planes.
I couldn't find similar explanations for planar and spherical datum features, but the logic could be extended to apply to those as well.
This new description doesn't explain how secondary and tertiary datums actually establish their theoretical plane(s), if any. But at least it releases us from the "two planes for a datum axis" and "three planes from a datum point" concept.
Evan Janeshewski
Axymetrix Quality Engineering Inc.
I can see how the center point is extracted from the two datum target points. The three mutually perpendicular planes associated with the center point still didn't make sense to me though. Part of the problem for me is that the "center point between two datum target points" does not have the same constraint properties as a center point extracted from a spherical datum feature.
One way that I evaluate my understanding of an issue is to imagine teaching it and dealing with the questions that would arise. Every class seems to have one or two students who just have to know where everything comes from and won't just take the instructor's word as gospel. If one of those students asked me to explain the logic of imagining three theoretical planes and then ignoring two, I couldn't.
So I read through the DRF chapter in one of the public review drafts of the new standard. The description of cylindrical datum features has significant changes.
1994: A cylindrical datum feature is always associated with two theoretical planes intersecting at right angles on the datum axis.
2009: A primary cylindrical datum feature is always associated with two theoretical planes intersecting at right angles on the datum axis. Depending on planes established by higher precedence datums, secondary and tertiary datum axes may establish zero, one or two theoretical planes.
I couldn't find similar explanations for planar and spherical datum features, but the logic could be extended to apply to those as well.
This new description doesn't explain how secondary and tertiary datums actually establish their theoretical plane(s), if any. But at least it releases us from the "two planes for a datum axis" and "three planes from a datum point" concept.
Evan Janeshewski
Axymetrix Quality Engineering Inc.
- Status
Similar threads
- Question
- Locked
- Question