【转帖】gdt a cross hole in a shaf

yang686526

gd&t a cross hole in a shaft
how would one dimension a cross hole in a round shaft? one method that was presented was to use a straightness on the cross hole and two perpendicular call outs to the axis of the hole, is there a preferred method of doing this? we have not been able to find any examples of this anywhere.
scorch
don't worry about people stealing your ideas. if your ideas are any good, you'll have to ram them down people's throats.
--howard aiken, ibm engineer

why not just positional?
perhaps an endface as -a- and the diameter as -b-, if orientation or 'clocking' is significant then add a -c- datum based on a feature to use for this.
fig 5-15 & 5-24 & 5-25& 5-36 in asme y14.5m-1994 kind of show this although they aren't great examples as they are all primarily showing something else.
are you iso or asme or what, if asme then you souldn't really use a cl as a datum in this application.
i'm not sure what the 2 perpendiculars get you in your proposed scheme.
kenat, probably the least qualified checker you'll ever meet...
if you are interested in having the cross holes in alignment with each other and also perpendicular to the c/l of the shaft, i would suggest that you draw phantom lines between the id of the holes. under the hole(s) size. place a feature control frame with perpendicularity to datum a which could be the od or id of the shaft creating a c/l.
both diametrically opposed holes are assumed as 1 hole and one can confirm perpendicularity by zeroing on one hole and then centre on the other. the difference should not exceed the perpendicularity tolerance.
one could go further by calling one end datum b. now we can place a basic dimension from datum b to the centre of the holes and the use a positional tolerance. we not only control the angle on both holes but their location.
just some thoughts.
dave d.
are  you in accordance with either the 1988 or 1994 version of y14.5.  i believe there is a significant difference with the interpretations.  
could you state the function of the holes?
why are you guys talking holes (plural). scorch talked about a (singular) cross hole.
i agree with kenat, just use a positional callout, but with the shaft diameter as datum "a" and an end face of the shaft or functional step (if there is one) as datum "b". the postiion tolerance will cover both centering on the shaft diametral axis and perpendiculity.
thanks for the responses guys, after i posted this the asme standard mysteriously appeared on my desk the next morning. kenat, i looked up the figures stated above and that showed me exactly what i was looking for. we will go with the positional tolerance. again thanks for the help.
scorch   
don't worry about people stealing your ideas. if your ideas are any good, you'll have to ram them down people's throats.
--howard aiken, ibm engineer

if it's relatively long in relation to it's diameter and one end doesn't really mate as a primary function, which is likely the case for a shaft, then i think rons datum order is probably correct.
most of my cylinders with holes are relatively short and mate at one end hence my suggestion.
kenat, probably the least qualified checker you'll ever meet...
yeh, application is everything. i recently did several stepped shafts with cross holes going through the larger diameter. (e.g. figs 6-47,6-48)the cross holes were for close tolerance bolts, locking the shafts into tight fitting pockets in bosses, so i was more concerned with alignments of the diameters than the distance from the bolt hole to end of the part. thus the shaft diameter as the primary datum.