huangyhg

assigning a datum axis through a tapered (countersunk) holes
folks-
how is this done? you can't create the datum to either the small diameter of the large diameter because the center of a circle is a point. on the other hand with a normal hole if you assign a datum to the diameter then you've got a valid datum axis where the center of the cylinder is an axis.
it was suggested to assign, say, c to one diameter and, say, d to the other. the axis would then be the compound datum c-d. the problem with that is that there is no precedent in y14.5m establishing a datum point!
this is such a common case in real life that it amazes me that y14.5m hasn't covered it!
tunalover
check out our whitepaper library.
not 100 % sure that i understand the problem. but, if i do, there are examples of using datum target points to get c and d and subsequently axis c-d. you could then relate the c-d to whatever feature you desired.
assign the datum to the conic surface.
editorial comment
a countersink seems like an awfully flaky feature to use as a datum. difficult to pick up with a gage, lots of variance in a small zone for cmm. why not use the hole and not the c'sink?
datum the c'sink hole. if there is not a hole, datum the bigger dia of the c'sink then dim the angle. i would not datum both dia's of the c'sink.
chris
systems analyst, i.s.
solidworks/pdmworks 05
autocad 06
correction: forget i said "countersink." the hole is tapered.
tunalover
because of chamfers and/or edge radii, the ends of a hole do not actually exist, so they don't make good datums.
i don't know what 14.5 has to say about it, but the traditional way of measuring the location of a tapered hole or using it as a datum is with a gage ball of specified diameter, the diameter usually being chosen so the ball stops halfway through the cone.
mike halloran
pembroke pines, fl, usa
thetick-
from a functional standpoint, hole centers are often good choices for datums especially for the control of coaxial features. it follows then that tapered holes often serve as natural and functionally-significant datums.
i've always held that functionality carries 85% of the weight when choosing datums! inspectability carries the remaining 15%.

tunalover
hole centers work well enough. my concern would be the quality of the c'sink surface. it would need to be a high quality cut to be used as a reliable datum. no chattering and well-centered on its respective hole.
mike h's gage ball idea would work well enough, though again depending on the quaity of the cut.
ringman-
where are these examples you mention?
tunalover
tunalover,
try fig. 4-34 on page 76.
ringman-
are you referring to y14.5m-1994?
tunalover

__________________
借用达朗贝尔的名言：前进吧，你会得到信心!
[url="http://www.dimcax.com"]几何尺寸与公差标准[/url]