hexagon logo
Register Sign in
  • Replies 11 replies
  • Subscribers 18 subscribers
  • Views 181 views
  • Users 0 members are here
Options
Related

Datum Precedence

KyleB
Hi all,

Today's conversation is datum precedence.

Should this be A|C|B and can it be A|B|C? (according to asme y14.5)

My understanding is that it SHOULD be A|C|B and that A|B|C is wrong - but it has now turned into a topic of debate as it cannot be found in the standard.
  • I don't think can/may/must can be found there either (maybe I am wrong) but that is the way it works.

    There are 6 degrees of freedom.

    If it CAN constrain a degree of freedom, and it MAY (because it hasn't been constrained yet) then it must.

    At a guess since I don't have your part, "A" looks like the center bore. CAN it constrain 2 degrees of rotation? YEP, none have been constrained yet. CAN it constrain two axial degrees of freedom? YEP, none have been constrained yet, so "A" constrains 4 of the 6 degrees of freedom.

    Next is "B", the end of the part. CAN it constrain 2 degrees of rotation? Yep, but may it? Nope, both have been constrained already. CAN it constrain 1 axial degree of freedom? YEP, the one it can constrain hasn't been constrained yet. So "B" does 1 axial degree of freedom. We are now at 5 DOF.

    That leaves "C", CAN it constrain 1 degree of rotation? YEP, and it hasn't been constrained yet, so it must. CAN it constrain any axial degrees of freedom? YEP, but it can NOT as all of them have already been constrained. So, it does 1 degree, giving us the 6 of 6 that we need.
  • PER GD&T ASME Y 14.5 2009 Paragraph 4.10
    (if you google for a pdf you can generally find a copy for free)
    4.10.1 Development of a Datum Reference Frame for
    Parts With Planar Surface Datum Features
    The feature control frame in Fig. 4-2 illustrates the
    datum reference frame for the part shown in its functional
    assembly in Fig. 4-2, illustration (b). Figure 4-2
    illustrates the development of the datum reference
    frame along with degrees of freedom. The datum features
    referenced in the feature control frame immobilize
    the part and constrain the six degrees of freedom (three
    translations and three rotations) to establish a datum
    reference frame. Relating a part to a datum feature
    simulator and a datum reference frame in this manner
    ensures consistent understanding of engineering
    requirements. See Fig. 4-1.

    (a) In Fig. 4-2, illustration (a), datum feature D is
    specified as the primary datum feature. Where a surface
    is specified as a datum feature, the high point(s) on the
    surface establish a datum plane. This primary datum
    feature contacts the datum feature simulator on a minimum
    of three points     (see para. 4.11.2 for discussion on
    rocking or unstable datum features). In this example,
    where the primary datum feature contacts the datum
    feature simulator, three degrees of freedom (one translation
    and two rotations) are constrained: rotation about
    the X-axis (u), rotation about the Y-axis (v), and translation
    in the Z direction.

    (b) Datum feature E is specified as the secondary datum
    feature. This feature contacts the datum feature simulator
    at a minimum of two points. See Fig. 4-2, illustration (d).
    In this example, where the secondary datum feature contacts
    its datum feature simulator, two degrees of freedom
    (one translation and one rotation) are constrained: translation
    in the X direction and rotation about the Z-axis (w).

    (c) Datum feature F is specified as the tertiary datum
    feature. See Fig. 4-2, illustration (e). In this example,
    where the tertiary datum feature contacts its datum feature
    simulator at a minimum of one point, the remaining
    degree of freedom     is constrained: translation in the
    Y direction.
    4.10.2 - 4.10.4 goes into more detail when cylinders are involved as datums, and how constraining should be conducted.

    The order of the datums should always control the most axii of freedom from most to last, in the specified order, as what said. -CAN MAY MUST-

    Another typical method , and in paraphrasing the bold portions of the GD&T Spec above, is by control via planar points of contact 3, 2, 1.
    -Three points equal a level plane, controlling three degrees of freedom,
    -Two points equal a line about the plane, and controls two degrees of freedom,
    -And one point to control the translation of the remaining axis of freedom.

    PCDMIS corraborates this during their training. All alignments within PCDMIS should be conducted by assigning LEVEL, ROTATE, and ORIGIN (Translate).
  • I'm not certain, but I think A|C|B and A|B|C accomplish the same thing on that part. As Matt says, Datum A constraines 4 DOFs. B and C can only each constrain one DOF not already constrained by Datum A.

    Either way, their use of datums A|B|C, in the order they defined, are what I'm accustomed to on round parts.
  • in reply to louisd
    so technically, ACB and ABC are both correct, and entirely interchangeable in your instance.
    Why? Because B and C only control one axis of freedom each. The outcome, mathematically, won't differ one bit, if you switch between ACB or ABC datums.
  • in reply to Matthew D. Hoedeman
    I'm impressed you whipped that all out in under 13 minutes.
  • in reply to Matthew D. Hoedeman
    Thanks for the very thorough response. I read this too but in less clear terms than you have spelled out. So if one degree of freedom constrains more than should it be first?

    also:

    "you whipped that all out in under 13 minutes"
  • It must just be some form of courtesy or rule of thumb that I'm feeling. I feel like you should always put the rotational datum first.
  • It must just be some form of courtesy or rule of thumb that I'm feeling. I feel like you should always put the rotational datum first.

    In my world, mostly turned and milled aerospace parts, the traditional alignment is a plane as primary, a turned diameter as secondary, and a timing hole as tertiary. Of course there are many other GD&T callouts used, but that is usually the most common one, which puts the third rotational datum last. Otherwise it would constrain translation degrees of freedom, which they don't want.

    It's really up to the engineer to use GD&T to represent how the part will be physically aligned in assembly.
  • Matt Hoedeman is spot on with the implication of the can, may, must rule and it is spelled out in the newest release of the standard. William Tandler taught this in his GD&T training for a long time even though it was spelled out in the standard.

    The DRF in a FCF should be based and what the design intent is (even though we all know that most design engineers suck at this).
  • My 0.02cents..

    Always level, rotate, XYZ in that order.

    Following CAN-MAY-MUST...

    Measure -A- as a cylinder with Z vector
    Measure -B- as a plane with Z vector
    Measure -C- as a cylinder with Z vector

    New alignment recalling startup...
    Level to -A-
    Rotate via offset line constraining -A- to -C- (can't rotate to -B- because it has the same nominal vector as -A-...so... we move on to -C-)
    -A- is X&Y origin, -B- is Z origin
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement