hexagon logo

Profile using a datum from a different part

I've just encountered a profile feature data frame on the part i'm checking where the primary datum is on a completely different part.
It's my opinion this is wrong and that all datums should exist on the part itself. However, i'm not 100% sure about this.
Parents
  • Also very possible that what matters is the profile of the surface ON the assembly.

    That should have been worked out during quoting (if you are not design), or during PFMEA if you are design. If you are in an industry that doesn't do that, basically, when desigining something like that, someone must make allowance for how things are going to be made to achieve the necessary dimensions on every assembly made. Like, machine that profile AFTER assembly, or slice the tolerances so summation of tolerance doesn't kill you on assembly, or have some tooling that simulates the assembly during manufacture and inspection. Something.

    Assemblies with dimensions and GD&T aren't uncommon, working the solution on how to make the components for assembly success is just something that needs to be done.

    So, maybe this is "OK." If they have a .010 profile callout during assembly and it would require calling a profile of .002 on each part in the assembly to net a good assembly, it might be better to call things to an assembly datum structure.
Reply
  • Also very possible that what matters is the profile of the surface ON the assembly.

    That should have been worked out during quoting (if you are not design), or during PFMEA if you are design. If you are in an industry that doesn't do that, basically, when desigining something like that, someone must make allowance for how things are going to be made to achieve the necessary dimensions on every assembly made. Like, machine that profile AFTER assembly, or slice the tolerances so summation of tolerance doesn't kill you on assembly, or have some tooling that simulates the assembly during manufacture and inspection. Something.

    Assemblies with dimensions and GD&T aren't uncommon, working the solution on how to make the components for assembly success is just something that needs to be done.

    So, maybe this is "OK." If they have a .010 profile callout during assembly and it would require calling a profile of .002 on each part in the assembly to net a good assembly, it might be better to call things to an assembly datum structure.
Children
No Data