TP Manual Calculation z,y,z

Hi guys,

Basically I'm measuring a bunch of parts for a capability study. Long story short I need to get true position on four similar features. For this instance I'm not sure how to accurately pull this off with the feature inside PC-DMIS since I'm trying to get it from 3 midpoints (1 in x, 1 in y, 1 in z). I figure I can use all the deviations and export the numbers into my excel and just formula it out. I would think the formula for a radius of a sphere would be the same for true position where TP would be the same as r. In that case r^2=x^2+y^2+z^2, or TP = SQRT(x*x+y*y+z*z). But on another forum post I saw, TP = SQRT(x^2+y^2++z^2)*2. What gives with the *2? They said that was for a circle? My shape is virtually a rectangular box. I don't understand how the middle of one object could be different from the middle of another object if it's "borders" were arranged differently?

Would, TP = SQRT(xdev^2+ydev^2+zdev^2), be the correct formula for getting TP on a rectangular box relative to called out datums?

Thank you guys!

Parents

frop1 over 10 years ago

There would also possibly be some allowable Datum Shift from C(M)

For this case C datum is the origin of the axis this thing wants TP of. And for once they give some unreal tolerance (in a positive way for a manufacturer). So my CPK value will be awesome even if it was MMC or RFS.

Okay well thanks alot guys for the help. I feel really dumb for asking these things but if I didn't ask I wouldn't learn. This will prevent this mistake on any other task. I also feel kind of foolish in that PCDMIS has the TP button which I can just select a single axis and yes it reports yDev and TP which is 2x yDev as everyone is trying to tell me. I'm going to read up on my free time to clear up some small questions in my mind. The good news is my 120 part measurement is all on a spreadsheet so I don't have to use PCDMIS to reload reports, can just formula it.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel

Reply

frop1 over 10 years ago

There would also possibly be some allowable Datum Shift from C(M)

For this case C datum is the origin of the axis this thing wants TP of. And for once they give some unreal tolerance (in a positive way for a manufacturer). So my CPK value will be awesome even if it was MMC or RFS.

Okay well thanks alot guys for the help. I feel really dumb for asking these things but if I didn't ask I wouldn't learn. This will prevent this mistake on any other task. I also feel kind of foolish in that PCDMIS has the TP button which I can just select a single axis and yes it reports yDev and TP which is 2x yDev as everyone is trying to tell me. I'm going to read up on my free time to clear up some small questions in my mind. The good news is my 120 part measurement is all on a spreadsheet so I don't have to use PCDMIS to reload reports, can just formula it.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel

Children

No Data