True positions of non-circular features

[DeWain Hodge]

I'm programming several parts that have a true position of a width with an MMC modifier. I have no problem measuring either side of the feature and getting the width. But the only way I know to find the true position is to ceate a midpoint and dimension the midpoint as a true position. The problem with this is that it is RFS and calculating MMC manually.

Any help would be appreciated.

DeWain Hodge

[rerun_lives]

Create a generic feature feature such as a circle with a nominal diameter equivalent to your nominal width. Assign a variable to the measured value of the width of your slot and use it as the "measured" diameter of the created generic circle. Then do a True position with mmc on the created circle. Use the same tolerance limits for the diameter of the circle as the width of your slot.

[DeWain Hodge]

Maybe I'm being dense but how does this refer back to the width? What about the vectors? Would the x,y,z be the same as say the centerline of the points I'm taking for the width?

DeWain

[kpayne]

Create a generic feature feature such as a circle with a nominal diameter equivalent to your nominal width. Assign a variable to the measured value of the width of your slot and use it as the "measured" diameter of the created generic circle. Then do a True position with mmc on the created circle. Use the same tolerance limits for the diameter of the circle as the width of your slot.

This seems interesting and appears it should work. I have some parts coming next week with slots that have true position call outs and I will have to give this a try.

[dwade]

Create a generic feature feature such as a circle with a nominal diameter equivalent to your nominal width. Assign a variable to the measured value of the width of your slot and use it as the "measured" diameter of the created generic circle. Then do a True position with mmc on the created circle. Use the same tolerance limits for the diameter of the circle as the width of your slot.

I do this a lot and it works great. If your feature is not square to the datums, assign variables to the IJK vectors and put them right into your generic feature.

[cmmguy]

Create a generic feature feature such as a circle with a nominal diameter equivalent to your nominal width. Assign a variable to the measured value of the width of your slot and use it as the "measured" diameter of the created generic circle. Then do a True position with mmc on the created circle. Use the same tolerance limits for the diameter of the circle as the width of your slot.

What do you do about the non-width axis. Using a circle would require that axis to be perfect or it would impact the MMC calculation. Do you use the measured value of the "other" axis as the nominal?

[dwade]

Measure the width of the slot at both ends. Use variables width or tolerance and report NONE. Create a generic circle using the vaiables of ALL of the characterstics of the mid and width points.

[robehor4]

example

Pnt44 =feat/point,rect
Theo/-1.3955,29.75,0,0.2588191,0,0.9659258
Actl/-1.3955,29.75,0,0.2607686,-0.0000393,0.9654013
Constr/point,mid,pnt40,pnt41
Dim Dist3
Ax Meas Nominal Dev +tol -tol Outtol
M 0.75000 0.75000 0.00000 0.00000 0.00050 0.00000
Pnt42 =feat/point,rect
Theo/-1.3955,35.5,0.375,0,0,1
Actl/-1.3955,35.5,0.375,0,0,1
Constr/point,int,lin23,lin20
Pnt43 =feat/point,rect
Theo/-1.3955,35.5,-0.375,0,0,1
Actl/-1.3955,35.5,-0.375,0,0,1
Constr/point,int,lin23,lin21
Pnt45 =feat/point,rect
Theo/-1.3955,35.5,0,0.2588191,0,0.9659258
Actl/-1.3955,35.5,0,0.2607686,-0.0000393,0.9654013
Constr/point,mid,pnt42,pnt43
Dim Dist5
Ax Meas Nominal Dev +tol -tol Outtol
M 0.75000 0.75000 0.00000 0.00000 0.00050 0.00000
Assign/rtkdia = Dist3.meas
Assign/ctrzrk = Pnt44.z
Assign/ctrxrk = Pnt44.x
Assign/ctryrk = Pnt44.y
Assign/ltkdia = Dist5.meas
Assign/ctrzlk = Pnt45.z
Assign/ctrxlk = Pnt45.x
Assign/ctrylk = Pnt45.y
F2 =generic/circle,dependent,rect,out,$
Nom/xyz,-1.3955,29.75,0,$
Meas/xyz,ctrxrk,ctryrk,ctrzrk,$
Nom/ijk,0,0,1,$
Meas/ijk,0,0,1,$
Diameter/0.75,rtkdia
F3 =generic/circle,dependent,rect,in,$
Nom/xyz,-1.3955,35.5,0,$
Meas/xyz,ctrxlk,ctrylk,ctrzlk,$
Nom/ijk,0,0,1,$
Meas/ijk,0,0,1,$
Diameter/0.75,ltkdia
Dim Loc12
Ax Meas Nominal Dev +tol -tol Bonus Outtol
Z 0.00000 0.00000 0.00000
Df 0.75000 0.75000 0.00000 0.00050 0.00000 0.00050 0.00000
Tp Mmc 0.00000 0.00100 0.00050 0.00000
End Of Dimension Loc12
Dim Loc14
Ax Meas Nominal Dev +tol -tol Bonus Outtol
Z 0.00000 0.00000 0.00000
Df 0.75000 0.75000 0.00000 0.00050 0.00000 0.00000 0.00000
Tp Mmc 0.00000 0.00100 0.00000 0.00000
End Of Dimension Loc14

[rerun_lives]

What do you do about the non-width axis. Using a circle would require that axis to be perfect or it would impact the MMC calculation. Do you use the measured value of the "other" axis as the nominal?

You are correct CmmGuy. I'm assuming that you would use one axis only if the walls of the slot are "normal" to the relative Datum Axis'. I haven't seen the print so am making assumptions. If the slot was floating or not normal perhaps it would be best to create a generic feature out of the corner intesections if it is a square ended slot. Also, I wonder if this is a rounded end slot. I didn't think about that......

[JIM HARRIS]

A little trick I use if Iam only looking for a true pos in one direction and it keeps it real simple is to take a vector point on 1 side of the slot or feature,then 2 points on the opposite side at the same loc.Create a circle from the 3 points then dim. your true position of the circle.

[cmmguy]

A little trick I use if Iam only looking for a true pos in one direction and it keeps it real simple is to take a vector point on 1 side of the slot or feature,then 2 points on the opposite side at the same loc.Create a circle from the 3 points then dim. your true position of the circle.

Wouldnt your circle be offset towards the "two point" side?

[JIM HARRIS]

No,your taking the 2 points at the same loc.Try it,it works fine.

[cmmguy]

The math for a circle wont work if two of the points are in identical locations. So that leaves me with other questions...