【转帖】he meaning of frtzf without datum reference

yang686526

the meaning of frtzf without datum reference
hi :
the books i read say the lower segment without datum reference controls the sapcing, the perpendicularity and parallelism are controled by the upper segment. i am confused and i really don't know what is the real meaning of this example.
thanks in advance for all inputs.
season

the lower part of the composite callout (in this case) is refining the coaxiality of the two in-line holes relative to an axis created by both holes simultaneously. the tolerance is specified rfs so no matter what size either hole is... each hole's individual axis must reside within a 0.01 diameter cylindrical boundary from an axis created by both.
if the tolerance was specified with a mmc modifier then a pin of dia. 0.123 could be used to verify the coaxiality along with separate measurments for size. the collout, however is not modified mmc so the each hole's short axis must be acquired and measured for orientational displacement from an axis created by both.
paul   
paul,
your explanation appears to be logical enough. however, it seems quite out unorthodox to have a positional tolerance expressed without referencing it to a single datum as a minimun.
are you not making an 'assumption" when you relate the holes to the axis created by both?
is there a similar example in the standard?  it appears that y14.5 is invoked on this drawing although there are a couple minor 'glitches'.

ringster,
it is not unusual or unorthodox to see a callout like this without a datum reference in the lower frtzf. see y14.5m-1994 paragraph 5.11.1.5 and figure 5-51. i have used these controls often to tolerance a series of coaxial hydraulic control valve bores.
it is unusual however, that the frtzf's tolerance is not modified "mmc" since the coaxial bores typically mate with a pin-like mating part. i suppose that if these holes mated with an expanding roll-pin and alignment of the holes were critical to achieve surface contact for retention then rfs may be appropriate... but i doubt that... some one probably just omitted the mmc because it wasn't on the previous design or they didn't consider its relationship to function or gaging.
paul   
ringster:
i agree with paul on this one.
one really needs the modifier mmc to be be able to confirm the alignment. one would then make a straight pin of the virtual condition size (mmc minus the tolerance) and insert it into both holes. the feature meets requirement if the pins are able to go into both holes with the force of no more than 1 finger.
without mmc, i really don't know how one would confirm the requirement.   
dave d.
paul & dave are correct, mmc makes most sense here.  you could verify the axis of each feature rfs and compare back to the coaxial axis between the two features, but there's not enough wall thickness to do this with great repeatability.  
the answer to "why" you don't need a datum reference on the frtzf is this; the pltzf (first level of the composite positional control) establishes the general zones based on the datum reference frame where the axes of each feature in the pattern must be located.  the second level (frtzf) is a refinement of the first lefel (pltzf), wherein the inter-feature positional relationship is being refined.  datum references are permitted on the pltzf if you are trying to simultaneously restrict the position of the features to each other and the orientation wrt your datum structure.
jim sykes, p.eng, gdtp-s
paul,
if we are to consider the 2 holes as a single axis, why would not perfect form at mmc be applied without the need for the lower segment of the composite tol?
that would be overly restricting.  envision the top level tolerance being a dia-1" zone; there are two coaxial tolerance zones which would be 1" dia, and therefore the axis of each feature must be within that zone.  now picture that on the second level, the tolerance zone is 1/4" dia.  that means that the two individual feature axes must be within a 1/4" dia cylindrical zone, but that 1/4" zone floats within the 1" zone.  
per ringster's proposal, i could tighten up to a 1/4" zone on the top level, but why make things tighter than they need to be?   
jim sykes, p.eng, gdtp-s
then we come around to another consideration.  what the heck is the assembled relationship?  that would be the determining factor, would it not?  otherwise we are guessing at the requirements.  
a similar exercise occurred in the 1994 version when resolving the old figure 142 of the 1982 version. imho.

ringster,
sorry for the late response i had to reply to dave... it took a while after church and such.
i disagree that it would be overly restricting as mechnorth stated. it is simply that the two holes are two seperate features and each one's "perfect form at mmc" is considered individually.
gm (in their uscar addendum) allows features to be delared "interrupted" which makes both one feature. outside of that... the coaxial position tolerance as was done in this composite example would be required to refine the coaxiality beyond the position tolerance.
paul   
i will restate my opinion. we need to know the assembled relationship to adequately evaluate the dimensional controls/requirements for this part.
i do not have at my disposal, currently, the 1994 version but the 1982 for coaxiality does ref a datum feature. fig 161.