【转帖】lever drf callou

yang686526

lever drf callout
lever drf callout
attached is a door lever we used popularly, the shank is assigned as datum b, there is a hole drilled and a square hole broached on the center. i am not making a test on you and want you to choose the right answer, but it's a real drawing callouts from our customer designer, different company designer will select different drf, i am confused on these drfs callout
a.for the central hole position callout
  * a drawing has three datum features is pretty common, shall we completely control the part (constrain all six degrees of freedom) for a position callout?
  * what is the meaning and differences between them if they are legitimate?
b.for the central square hole position callout
  * the diameter symbol is not allowed here and it have been discussed many times on eng-tips forums, but as a manufacturer they will request more tolerance as the cylindrical tolerance zone has, any other gd&t tolerance can meet this target?
  * same question again, shall we completely control the part (constrain all six degrees of freedom) for a position callout? is the callout correct on circle #6?
  * one of the print without datum a, only datum b and c shows on the print, is it legal on the position callout |pos|.010|b|c| as shown on circle #7?
c.for the central square hole perpendicularity callout
  * is ? allowed on perpendicularity callout here?
  * i can understand the meaning of |perpen|.004|a|b|, but what's the meaning of |perpen|.004|b|c|?
d.for the central slot position callout
  * does this position callout can control the perpendicularity with respect to datum c?
thanks for all comments and inputs
seasonlee

4doffff">---functionally speaking... my experience with door handle assemblies (installing occasionally and using daily) lead me to respond (with what i understand about geometric features, their assemblies, and how they constrain degrees-of-freedom)that the functional primary datum feature here is a cylinder "b" constraining four degrees-of-freedom. if so... there are two remaining degrees-of freedom left to constrain rotation about the axis and translation along the axis.
1doffff">--there is a groove for a (cir-clip "i suspect") on the cylinder "b" and "i suspect" that the width between the cir-clip and the surface identified as "a" is probably the feature that functionally stops translation along the axis.
1doffff">--there is also a broached "square internal feature" for which there is required a drilled "pre-finish" hole to process that "i suspect" is the other functional datum feature stopping rotation about the axis.
problems... considerations... choices...
primaryfff">
3dof "a" vs. 4dof "b
which clearance is less??? which one wins in constraining more degrees-of-freedom??? it depends upon how tight or loose "b" fits into the latch assembly...compared to how tight or loose the width between the cir-clip and opposing surface fits into the latch assmbly? the tighter one wins since the length the cylinder "b" is almost identical to the diameter of the surface identified as "a" if the cylinder wins it is primary and constrains 4dof... if not the width wins and it is primary and constrains 3dof.
secondary
if the cylinder won over the width then the selction of the secondary is abitrary between the width and the broached internal square feature because both only constrain 1dof.
if the width won over the cylinder the the cylinder must be the secondary because the cylinder (capable of contraining 4dof) will constrain all remaining degrees of freedom except the one that it cannot (rotation about its axis).  
tertiaryfff">
1dof "functional" vs. 1dof "repeatable... measurable
the internal square feature is functionally the datum feature. it  stops rotation about the axis rather than the end of the handle identified as "c" but... its width compared to the distance between the axis of "b" and the end of the handle "currently identified as "c" is substantial.  although the handle ends up where the small square broached surface puts it functionally... not the other way around as it is currently detailed... trying to measure (constrain) the variability of where the handle ends up in relation to the square would be subject to enormous error amplification compared to using the handle to stop rotation. that is why "i suspect" the drawing as it is now has the end of the handle identified as "c". if the end of the handle is ultimately chosen as the tertiary i would identify a couple of target locations on it near its end rather than relying on the intersections of cad perfect geometry.
paul
     
     
first, you need some controls to be established for datum features a, b, & c.  assuming that this is a simplified drawing for the purposes of this question and these controls would otherwise be in place, i'll make some assumptions about the functionality of this part:
- the handle is to sit flush on datum feature a and mate cylindrically on the boss datum feature b, then there is a desire to have the handle oriented along its longitudinal center plane
a. for this workpiece,
(1) wrt b only would not provide any orientation relationship to datum a;
(2) the best of the three options. wrt /a/b/, this provides a perpendicularity requirement wrt datum a, and a location requirement wrt datum b; as this feature is cylindrical and centered on your datum feature b, orientation effects wrt datum-c are not a concern.  
(3) adds reference to datum-c, but you're not concerned about rotational orientation here. on the other hand, if all other features use drf /a/b/c, then using the same drf will mean one less setup for inspection.
b. for the square broached hole, are you intending to control the position and orientation in the one direction only?  currently you have it shown for the center plane established by the vertical width of the broach only.  what about the other direction?
- you'd want a positional callout in both the horizontal & vertical directions with a linear zone
- a cylindrical zone wouldn't be legal here as you're dealing with opposed planes.
- if you want to maximize tolerances, look into an mmc modifier on the tolerance values for both the cylindrical and square feature
- you would want to have the position wrt /a/b/c/ as the orientation and location to each is important, so (4) is the best choice
- #5 is not legal because of the cylindrical tolerance zone
- #6 does not require an orientation to datum-c, therefore the square profile could be rotated about the b-axis and still be accepted
- #7 eliminates all reference to datum-a, so the broached hole does not have to be perpendicular to datum-a
c. drf /b/c/ does not relate back to datum-a so you don't get any orientation relationship back to datum a. if you visualize the simulator, you will have datum axis-b as primary with its two intersecting mutually perpendicular planes, and datum-c as secondary locking down the spatial orientation of those two planes.  so, drf /b/c/ would be technically legal, but wouldn't get you the relationship to datum-a.
- once you establish a new drf, your setup changes.  if you have /a/b/c/ for the position control, then /b/c/ for the perpendicularity refinement, you are no longer invoking datum a, therefore in your inspection setup, you have to make primary contact with datum-b, secondary with datum-c.
d. there is a hierarchy within gd&t wherein certain controls automatically include other controls.  profile of a surface, for example, when datums are referenced will provide a surface control for location, orientation and form.  a position control includes location but also orientation of the axis or center plane, so yes /a/b/c/ per the fcf does provide an orientation wrt datum-c.
jim sykes, p.eng, gdtp-s
since i didn't answer any of your questions in my reply but rather opined on function... my recommendations for your controls (unsolicited) would be:
considering that the fit of the cylinder b with its mating parts in the latch assembly wins fff"> over the fit of the width between the surface and the groove...
label the cylinder "a". don't bother with a form control for it since its form must be perfect (0) @ mmc and it only has a .002 tolerance for size.
label the width between the surface (currently called "a") and the groove wall opposing it... "b" you can tolerance each of them for their perpendicularity or runout to the cylinder axis "a".
(note the variability of the cir-clip size and compressible form + the width tolerance of "b" + the width of the mating latch assembly will determine its fit clearance/interference)
label two target line contacts (c1 and c2) at some basic distance from the axis of "a" and ... put "bird-beaks" equalizing symbols (found in the asme casting standard and included in the new y14.5 standard)on the target locators so that the width can be treated rfs.
for the square broached internal... (this is a hard one because i know that functionally its centrality to the cylinder "a" is not as important as its aesthetic "looks" orientation to a|c )... i would control its size tightly as well as that of the square rod to limit slop. then i would use a |position| (no dia.) tolerance (m)| a|b|c| (no datum modifiers).   
one could use a composite position (no dia. symbol) @mmc where the upper controlled its location to the cylinder "a" but... and the lower refined the tolerance for rotation but if both were specified |position| (no dia.) tolerance (m)| a|b|c| with different tolerance values one beneath the other sharing only the position symbol... most people would look at it and consider it a mistake (see figure 5-29 asmey14.5m-1994).
paul
in the lower i would reduce the tolerance for orientation  control  and if we use a composite that  
thanks for the comments.
paul
it seems you are quite familiar with the door lever and locks, as what i know there is a c clip to hold the lever on the groove as you said. a square pin will insert into the square hole, it will break automatically at the small neck area when a heavy force apply on the lever, it's a safety design to protect the lock and door, therefore i don't think the position is critical here.
there is another metal part sit inside the slot to control the rotation, so if i were the designer i will make the following changes:
* assign the slot as datum c. (as shown on the attached)
* change the shank as primary datum a (same as what you thought).
* give a tolerance on datum d to make it perpendicular to the slot datum c.
jim
your interpretation on the various drf callout makes me feel more confident on what i am thinking , thanks for your highlight on the mistake of square hole position tolerance callout, most people think the square dimension applies to both direction, so its tolerance will apply to both direction as well.
seasonlee