What’s a datum reference frame?
As discussed in workshop No. 5, datum reference frames (DRF) are coordinate systems, and preferably—at least to start with—Cartesian coordinate systems.
As we know, coordinate systems serve to orient and locate objects, and in the world of GD&T in particular, to orient and locate tolerance zones.
Focusing on the encircled feature control frame in figure 1, below, it’s common to think of the cylindrical tolerance zone defined by the position tool as being oriented and located relative to datums A, B, and C. That’s sort of right, but in fact, datums A, B, and C don’t act directly, and instead serve to establish a DRF ABC, which in turn serves to orient and locate the tolerance zone. Our purpose in this workshop is to present the orderly process for establishing DRFs.
|Figure 1. The part drawing|
This process, not described in the Y14.5M 1994 standard, is presented in some detail in §4.7, pp. 18-20 in the Y14.5.1M 1994 standard. SmartGD&T breaks it down into six easy steps, which are based on two sets of rules. The six steps and the most important (but not all) associated rules are listed below:
Datum Reference Frame
Establishment Process Steps
These rules and processes are part of our attempt to simplify and systematize GD&T to the point at which it can be encoded and decoded, rather than used to decorate drawings for later interpretation. Some of these rules and processes are explicit in the Y14.5 and Y14.5.1 standards; some are implicit, and some are missing. But in the end, every rule and every step is essential, and all are based on fundamental reality. In particular, the processes outlined above fully duplicate those performed millions of times a day in machine and inspection shops all over the world, in which datum features are turned into manufacturing fixtures (datum feature simulators), whose coordinate systems are then made to coincide with those in machine tools, and then transferred to actual parts by mounting them in the fixtures.
Making it happen
With this introduction, we’re ready to proceed with the process.
Step 1. Decoding the feature control frame
Position requires the bounded axis of the considered feature to lie within a cylindrical tolerance zone of diameter 0.5 mm at maximum material condition, expanding, which is oriented and located by basic dimensions relative to a datum reference frame established using datum feature A simulated rocking, datum feature B simulated stably regardless of its size, and datum feature C simulated mobly at its virtual maximum material boundary.
Aware of the requirements, we’re now ready to proceed to Step 2, where we identify the datum features so that we can create their inverses, namely the dimulators we’ll need in Step 3.
Step 2. Identifying the datum features
Datum feature A is a planar surface. Datum feature B is a bore. Datum feature C is a slot.
Step 3. Constructing the simulators
As shown in figure 2 below, the simulator for A is a “perfectly’ flat planar surface pointing in the opposite direction. The simulator for “B” is an expanding cylinder (expanding because the tolerance zone size modifier (S) requires it to consume all the available space). Finally, the simulator for C is a tombstone fixed in thickness at the width of the material-free zone inside the slot, namely at the width of its virtual maximum material boundary, equal to the minimum slot width of 16.5 mm, minus the width of the position tolerance zone of 0.5 mm, for a total of 16 mm.
Figure 2. The datum feature simulator set
Step 4. Extracting the datums from the simulators
Based on the definition of datums presented in Workshop No. 5, datum A is the tangent plane on datum feature simulator A, datum B is the axis of datum feature simulator B, and datum C is the midplane of datum feature simulator C, as illustrated in figure 3.
Figure 3. The datums and their associated datum feature simulators
Step 5. Using the datums to constrain a starter coordinate system in the simulators, thereby turning it into the sought after datum reference frame
In this process, datum A will constrain the pitching and yawing, as well as the translation in Z of the starter coordinate system, datum B will constrain two more degrees of translational freedom, bringing the Z axis in line with the axis of B, and datum C will constrain roll, the last degree of freedom, producing the DRF lodged in the simulators illustrated in figure 4.
Figure 4. The datum reference frame established in the simulators
Step 6. Marrying the datum features to their simulators to transfer the datum reference frame to the actual part
Marrying datum feature A to its simulator constrains pitch and yaw as well as translation in Z, but with some potential residual mobility. This is due to the Y14.5M 1994 rule of rocking datum feature simulation in combination with the possibility that datum feature A is somewhat convex. Marrying datum feature B to its initially unexpanded simulator, and then expanding the simulator to achieve stability, constrains translation in X and Y. Finally, marrying datum feature C to its tombstone simulator constrains roll, but only partially, depending on the extent to which datum feature C departs from its virtual maximum material boundary. The end result is illustrated in figure 5.
Figure 5. Transferring the datum reference frame from the simulators to the part
Because the whole purpose of establishing a datum reference frame is to constrain a tolerance zone, we now see that the cylindrical tolerance zone defined by the position tool is constrained in location, but is still free to pitch and yaw slightly relative to A, and roll slightly relative to C.
As described above, all this mirrors the process of building a manufacturing fixture, inserting it into a machine tool, and ultimately mounting a part in the fixture to cause its coordinate system to coincide with that in the machine. It also mirrors the process which coordinate measuring machine software is supposed to, and hopefully follows.
In retrospect, we now understand for the first time the intimate connections between machine parts and people, namely, that whether we’re marrying machine parts to their simulators or guys to their gals, the process is the orderly (well, most of the time) and sequential reduction in the degrees of freedom between the partners.
Our next SmartGD&T workshop will demonstrate the encoding process, the process of researching the function of a part and of its individual features, and then encoding those functions for later analysis in design, and, finally, for transfer to the machine shop and quality assurance. Subsequent articles will be dedicated to converting the code into a manufacturing process, and then into an inspection process.
Feedback on topics of interest
We got a couple of e-mail inputs in response to our last article, for which we were very grateful. Thank you Eddy Ang and Tim Schroeder. We really appreciate questions, so we can be sure we are addressing problems of interest.
About the author
William Tandler is the founder of Multi Metrics, a provider of geometric dimensioning and tolerancing technology and corporate implementation services. Through its SmartGD&T technology, training, and consulting products and services, Multi Metrics enables companies and individuals to realize the promise of GD&T.