In this Topic Hide
Structuring your Measurement Routine for Geometric Tolerances
Comparison to Past Practice - Constructed Features
Comparison to Past Practice - Measured Data in Cross Sections
Comparison to Past Practice - Considered Features and Measured Data
In most cases, for your measurement routine, we recommend a simple structure like this:
Construct an initial alignment to find the part in 3D space. For information, see the "Creating and Using Alignments" chapter.
Measure all the considered feature surfaces and datum feature surfaces.
Define the datums using datum definition commands.
Define the specified size tolerances and geometric tolerances using geometric tolerance commands.
Define any simultaneous tolerances using simultaneous tolerance commands.
You need to create the size tolerances and geometric tolerances on datums before any geometric tolerances that reference those datums. This is because geometric tolerances that reference a datum frequently need to know about any size tolerances and geometric tolerances on that datum. If you later edit any size tolerances on a datum, please ensure that all subsequent geometric tolerances that reference that datum have the correct size tolerance information for the datum.
We do not recommend using copy/paste or paste-with-pattern to replicate the geometric tolerance command. There are some cases where it works, and there are other cases where paste-with-pattern is fundamentally incapable of working correctly. We also do not recommend putting the geometric tolerance command inside a loop, for similar reasons: some cases work, and some cases cannot ever work. It is okay to put the entire measurement routine inside of one loop.
In the past, because of XactMeasure limitations, you often had to use constructed feature commands. These were mid-planes, intersection lines, and so on. You used them as either considered features or datum features.
With Geometric Tolerances, however, most constructed feature commands get in the way. Constructed features can impede a Geometric Tolerance from understanding the measured surface. However, in these few cases it makes sense to use a constructed feature command:
Constructed width commands are necessary to represent a width (ASME Y14.5) or opposed parallel planes (ISO 1101) feature. This is because there is not yet an auto-width command in PC-DMIS. Constructed width commands keep all the surface data, and so they do not impede the geometric tolerance command.
Constructed set commands are sometimes necessary. When all the inputs represent measured surfaces, constructed sets keep all the surface data, and so they do not impede the geometric tolerance command.
In rare cases it makes sense to measure a feature with separate point commands like vector points. You can then construct a Best Fit Recompensated (BFRE) feature from the vector points. The constructed feature retains the surface data, and so this does not impede the geometric tolerance command.
In rare cases, a considered feature or datum feature might need to be a derived geometry. This means it has no surface. An example of this is the minimum circumscribed circle that contains three pins in ASME Y14.5 2018 Figure 7-42 (b). In such cases, the only way to convey the specified intent to the geometric tolerance command is to construct a feature that lacks surface data. If you do this, you are responsible to comply with applicable standards.
The geometric tolerance command uses constructed features without surface information in these ways:
As the toleranced feature
As the pre-resolved datum feature (situation feature in ISO 5459 language)
In such cases, you are taking over phases 2 and 3 of the conceptual evaluation process. It is your responsibility to construct the feature according to appropriate standards. For information about the conceptual evaluation process, see "Introduction to Geometric Tolerances and Feature Control Frames". For information about which feature types have and do not have surface data, see "Feature Types With and Without Surface Data".
Certain types of geometric tolerances need measured data in cross sections. For example, consider the straightness of a cylinder’s axis. You need to measure the cylinder in several cross sections for these reasons:
To compute the center of each cross section
To evaluate the axis straightness of the circles’ centers
The XactMeasure command required you to measure several circle commands. These may have been child features of a scanning strategy. You then had to construct a 3D Best Fit (BF) line through the circle centers. Finally, you created an XactMeasure axis straightness tolerance on the BF line.
With the geometric tolerance command, you no longer need to go through so many steps. Now, you measure the cylinder in cross sections (using a measurement strategy or not). You then evaluate axis straightness on the cylinder. The geometric tolerance command automatically divides the data into cross-sections. It then computes each circle’s center and evaluates straightness on the centers.
With the geometric tolerance command, every considered feature has a measured value. In some cases, this is different than XactMeasure's behavior.
For example, take a surface profile tolerance of three circles. In XactMeasure it yielded only one measured value. But with a geometric tolerance command, it now yields three measured values.
This means if you only want a single measured value, you need to construct a set of the input features and make the set the considered feature.
We chose this new behavior because it gives more flexibility in how to report measured values. Single measured values are still available with a constructed set. But separate measured values are also available where they weren’t before.