【转帖】positional similar to asme y14.5m fig 4-8

yang686526

positional similar to asme y14.5m fig 4-8
i鈥檓 helping an intern dimension a part which has a datum scheme similar to figure 4-8 of asme y14.5m-1994.
say the 9.2 hole has a couple of 鈥榢ey鈥?(male) features spaced 180掳 in line with b-c.
to dimensions the keys i was going to use a scheme similar to 4-6.  to do this i鈥檇 make the 9.2 hole datum d, to ensure they are radial to that hole.
1.    does this make sense, it has fairly tight tolerance/functional requirements so i think this may be better than relating it back to abc.
2.    should my tertiary datum (for orientation) just be b or should it be b-c?  i鈥檝e looked at 4-9 & read the paragraphs and am leaning to b-c but want to make sure i鈥檓 right.
i know i鈥檝e been asking a lot of questions lately, probably more than i鈥檝e been answering but i appreciate the help.  (one of my departments goals for the year is apparently to give everyone at least 1 training so i鈥檓 hoping mine will be in slightly more advanced gd&t so i can quit bugging you all so much!)
kenat, probably the least qualified checker you'll ever meet...
check out our whitepaper library.
i'm tired, and have evidence already on eng-tips that i shouldn't be answer questions in this state, but i'll give a tiny bit of input that may or may not even be close to what you are asking.
1.  if you wish to emulate fig 4-6, then yes.  otherwise, i don't know cuz i don't know the intent/fuction of these features.
2.  4.5.7 says it all.  well, not really.  it differs back to 3.4.2, which establishes a pretty strict rule about using multiple datum.  i don't know if there are two different instances where to use them, or if one is the rule, and the other is the application.  i know this isn't what you asked.  i guess i'm kinda asking for what it meant in the standard for multiple datum features.  however, if i'm reading 3.4.2 right, b-c wouldn't be allowed since b-c consists of two features base on surfaces that have nothing in common, unless the new datum formed from those datums is the plane created by their two axes.  am i off base here?  i think i'll be learning from anyone's answer.
matt lorono
cad engineer/ecn analyst
silicon valley, ca
have you considered a profile on the hole/key feature?
thanks,
the keys locate into grooves into a cylindrical item with key-ways to orient it.
i don't see how 3.4.2 forbids b-c in this case, it's mainly talking about how to show it in the fcf.  i essentially want to invoke the datum plane c shown in the bottom left of 4-9.
profile might be an idea, although the scheme as is works quite well except that i want to make sure i have the orientation properly addressed.
kenat, probably the least qualified checker you'll ever meet...
either composite profile or profile w/position (fig. 6-19) should more than achieve what i believe your looking for.
sketch for clarification.
very simplified sketch of part showing the datums.  there are a bunch of other features too.
b & c locate on dowel pins and align this part to other, a mates to the others.  there is another hole patter that relates to d which i haven't shown.
the keys & diameter are critical, the small corner releifs aren't.
thanks for the help.
kenat, probably the least qualified checker you'll ever meet...
kenat,
   i do not see a problem with adding a fourth datum.  if you need to locate something to your 9.2mm hole more accurately than to the main datums, then you should specify the hole as datum_d.  datum_d provides centreing.  i would use datum_b or datum_c to control rotation.  perhaps someone out there can visualize a fixture that uses b-c to control rotation, without the additional location control.
   it is hard to go any further than this without understanding what your part does.  there are all sorts of valid approaches.  what must they accomplish?
                          jhg
thanks drawoh, i answered 1. for my self by looking through some training material i had and am confident in adding the datum d.
it's q2, getting the keys aligned correctly i was struggling with.  i wasn't sure just b (or c) defined a datum for this.
kenat, probably the least qualified checker you'll ever meet...
ok, i would go with position to datums a,d, and b-c for clocking, for your keys. hole "d" probably being positioned relative to a,b, and b-c.
the tolerance values look hugh, but they are arbitrary numbers?
if the dowel holes share equal function, use of pattern datum is the most robust and encouraged.  make the pattern "b", addressing your secondary/tertiary datum and locking the part.
the feature in question here is not a hole it is commonly referred to as an "irregular feature of size", while others may call it a bounded feature.  in some respects it reminds me of a spline, but we won't go there.
i sense some hesitancy in applying the profile(all around) combined with position and invoking "boundary" and even possibly @ mmc.  i would imagine that this is because of the corner reliefs.  if this is the only reason, look at para 6.5.3 and fig. 6-12.  as you can see all you will need is a local note, overriding the spec in regards to the reliefs only.
now once you apply this aforementioned callout, re  
after reading what xplicator has just posted, i like it and would consider it, especially if the tolerance size tolerance on the hole and keys needs to, or can be the same. it may take a little verbal ustification, but should be clear.