【转帖】olerance analysis

yang686526

tolerance analysis
hey guys,
   i have a problem which i believe is simple but i am having a hard time figuring it out in my head.  basically what i have are two pins (.018 ±.001).  
these pins gets inserted into tooling which have holes of .020 ±.001 positional of .004
so now i have these pins set in the tooling.
a flat piece with holes (.033 ± .003 positional of .005) in it gets placed over the pins
so far i am calculating that i have (.030 - .019 = .011 - .005 - .004 = .002 / 2 = .001 <--- the distance that the two holes are from being tangent in a worst case.
now the pins get soldered into the flat piece of metal and it is taken out of the tooling.
this assembly (the flat piece of metal with the soldered pins) needs to be assembled into another piece of metal with holes in it (.033 ± .003 positional of .010 l).
this is where the problem comes in, the holes in the second piece of metal have such a large positional @ mmc that i don't believe the assembly could work.  the only thing i can change is the tooling hole tolerances and the first piece of metal tolerances.
in my head i don't believe the positional tolerance on the first piece of metal will help me at all (to an extent) because the tooling really decides where the pin sits, the first piece of metal is really just a carrier, the pin just gets welded in wherever.
i see it as unsolvable in that due to the nature of the pin size, even if you had a positional of 0 on the tooling hole it still wouldn't work.
help...
2x .018 ±.001 pins going into 2x .020 ±.001 holes with .004 positional tolerance - assuming perpendicularity of 0 on soldering operation - yeilds a virtual condition of .025 (.020+.001+.004)
2x .033 ±.003 with .010(lmc?) - even dropping the lmc, if that is what was meant, yeilds a virtual condition of .020 (.033-.003-.010). an lmc would only aggrevate the scenario.
a .025 v.c. pin going into a .020 v.c. hole is going to provide a transitional fit i.e.: some will slip fit, others will be a press.
weavedreamer,
   i appreciate the response and it seems that i was right when i said that the first piece of metal is meaningless because the second has a looser tolerance.  there is a lmc on the .010 so it would be (.033-.003-.016 = .014.  
thanks again,
pete
looking at your response again, where does the pin size come into play in your calculation of the virtual pin size.  i get that the pin can move around in the .020 ± .001 hole but shouldn't the pin size itself come into play.  what am i missing?
the pin can move abour in .020 hole - giving it the .020 diameter.
weavedreamer and pruggiero,
   you guys are assuming that the pins sit perpendicular to the two plates.  actually, they lean over a bit, increasing your virtual condition.
   how deep are the holes the pins sit in?  how perpendicular are these holes?
     
               jhg
drawoh,
   none of the holes have a perpendicular tolerance on them, does the positional take care of this or not since it isn't a projected tolerance?  both metal plates are about .060 thick.  i see weavedreamer point of the pin virtual condition, this is because the virtual condition is worst case a smaller pin size doesn't affect the virtual condition but it could help.

pruggiero,
   your positional tolerance allows the hole to run from one edge of the positional diameter, to the opposite edge.  if you need perpendicularity, you need to specify it.
   even if your holes are perpendicular, the pin can still slope from one side of the hole to the opposite side.  a small pin will slope more than a big one.  a ?.017" pin in a ?.021" hole has .005" space to lean over, with a height of .060".  that works out to around 5°.  the virtual condition .060" above the first plate is around ?.035", not accounting for the hole's perpendicularity, tolerances in the length of the pin, and chamfers of the hole.
               jhg
pruggiero,
can you show a pic of the drawing?
chris
solidworks/pdmworks 08 3.1
autocad 08; catia v5