【转帖】hole locations for holes to be match drilled later

yang686526

hole locations for holes to be match drilled later.
if the task was to dimension a rectangular sheet of aluminum with pilot holes to be match drilled at subsequent assembly, and geo tol were applied, which would be the preferred locational tolerance for the pilot holes and why?
diametric or square (rectangular)?
any thoughts on this?

the pilot holes should follow the same scheme as is necessary for the final assembly. in my little world, there are very few cases where i would choose a rectangular position over a diametric, simply because there is better axial alignment in the diametric.
v
ringman,
   if i am specifying holes for screws and bolts, the error that will get me in trouble is a radial deviation from nominal.  my optimal tolerance zone is circular.  a rectangular tolerance zone that accomplishes the same thing will have a much smaller area, and it will reject parts that will work.
   if you are fabricating one set of holes and using these to pilot the mating holes, then mating the bolts does not matter.  the next requirement is the location of your bolts.  
   is there anything the bolts might interfere with?  
   how much like a piece of crap is it acceptable for your assembly to look like?
   i think your tolerance zones still should be circular, but it really depends on your requirements.  
                      jhg
i'm with vc66 and drawoh. stick with circular position tolerance, especially for pilot holes which locate the final hole. the machine will work in x,y coordinates, why restrict your tolerance range to 57% of available?  
if we were to think outside the circle and apply a square zone to the pilot holes, it would give the 'bonus tolerance'
if the square were circumscribed about the circle.
it would not in any way affect the accuracy of the hole size to be made at the rivet installation  dwg,
nor would it violate the edge distance allowance.
well, it is a pilot hole and may bias anything that comes after it.  i would first consider positional (which does provide its own bonus tolerance zone).   
matt lorono
cad engineer/ecn analyst
silicon valley, ca
ringman,
   the usual reasoning with positional tolerances is that the square is circumscribed inside the circle.  the circle provides the larger area of acceptable positions.  this all assumes that your maximum allowable error is a radial deviation from nominal.
   what are you trying to accomplish?
                    jhg
the right choice would be circular (but see below).  
just a thought on this subject - why does it matter?  if the holes are the datums disregard the following.
i'll explain my thinking.  the holes will be redrilled and the material is alum (which is soft).  so if the hole sizes are not provided the work around would be to give a position of .030 sq or cir (a good liberal tolerance) which would sufffice, as long as you leave enough stock to redrill.  then niether one would would be a problem.  plus you would save time inspecting the holes before assembly (keeping it off the cmm).   
my thinking was to use gd and t and at the same time maximize the allowable tolerance for the pilot holes.
i believe the round tolerance serves the purpose to insure interchanceability of screws and such, but not mandatory for pilot holes.
i hve also advocated the use of square for the pltz for composite tolerancing.
ringman, you've brought up similar several times before.  generally i'd say circular is preferable for the reasons others give above.
however, i'd never say never about the rectangular zone, if that's what the function requires then so be it.
if in your case you're more concerned about the holes not being to close to say a straight edge rather than their location relative to each other then maybe rectangular would work for you.
kenat, probably the least qualified checker you'll ever meet...