Position of Sphere to Plane

[JacobCheverie]

I have a part with a planar datum (width). The part has an I.D. sphere that is supposed to be on the datum plane (basic = 0). The print has a Ø tolerance zone modifier.

I would assume that the sphere (point) can only be measured to the datum in one direction (normal to the plane) and therefore the print should have no tolerance zone shape modifier (planar zone). Am I correct?

Further, the width has a MMB modifier on it. When I measure to a planar zone, no datum displacement is occurring to better optimize the Position result. When I switch to a diametral zone, datum shift occurs.

[SingularitY]

I've seen a lot of mistakes on prints where diametric symbols were added by mistake. Force of habit. Can you provide a snippet of the callout?
If the callout is truly intended to be planar, then it would only have a deviation in one direction, you are correct.

[sealevel]

Or.... maybe the designer forgot (SP)Ø in the callout AND forgot a secondary datum....
yeah, i know.... that would be a lot of forgetting

[JacobCheverie]

SingularitY I have attached a snippet of the callout.

As you can see, A is a width and B is an I.D. sphere. Position on B is called out diametrically at MMC to A at MMB.

TEST.jpg

[KIRBSTER269]

Late to the game on response on this sorry, A sphere is quite different from a cylinder, it's kinda of like a circle, but a 3D circle. What I mean by that is that a sphere is controlled by a common center POINT. So you have to treat it as a Radius instead of a diameter. As it is showing a position callout is to "A" with the FCF, you have to establish something, in this case 2 planes creating a width finding a common center plane, This is why I believe that MMB for A is pretty useless, and now creating a sphere on that common plane. To control a sphere all around, you have to have a common center point. Now does that common center point lie on that plane. So with the MMC on the sphere, would allow slop in all directions, Matt states perpendicular, I look at it as parallel to "Z" problem is how do you check parallel or perpendicular to a point? No matter how big that sphere is, or how small, it is controlled by a center point, creating a Polar radius in all directions, allowing rotation of in this case, your picture, let's say "Z" , with MMC which direction am I allow to Shift? Don't know because there are no other features controlling "X" or "Y" So does that Common point lie on that common plane created from the width? This callout is calling out to A, your picture to the right, your sphere needs to lie on any location in the "X" or "Y" direction of that established "A" so shifting in any direction in "X" or "Y" is allowed. but there might be other callouts to prevent this.

[SingularitY]

I'd dimension this a planar. If you really want to, you could dimension it as spherical but 2 directions would show perfect.

[JacobCheverie]

Thanks, that's what I assumed as well. The issue(?) is that with a planar zone, absolutely no datum displacement is occurring to optimize my result. It is only when I switch to a 2/3D tolerance zone that shift starts being applied. Surely I should see some X/Y Rotation or Z Translation of datum feature A to minimize Positional error, no?

[SingularitY]

I wouldnt know why datum shift isn't happening without seeing the code involved. It matters the size of datum A. I wouldn't expect any kind of rotational shift since the spherical feature itself is the origin in 2 directions, so any angular shift wouldn't help the position any. (It's your pivot) It should translate in one direction though, unless the width is at maximilum size or above. The customer basically wants to make sure this feature is in the center of A and calculating bonus for both the feature and Datum A would be simple since no rotating shift is involved.

[Matthew D. Hoedeman]

I can foresee a lot of issues trying to dial that small section of sphere to the plane, not really enough sphere there for real accurate values perp to the plane. And, since the sphere is "B" it will be zero in 2 axis, the only deviation will be perp to the plane. Reporting a position will make it hard on the machinist to tweak the location in. You could simply report the "Z" axis (if you level to Z) as a +/- value/tolerance, with the tolerance value being a variable of the MMC of the sphere.

[JacobCheverie]

Z = 0 +/- (TOL + ØACT - ØMMC)/2?

Wouldn't the Xact Position report the same thing and give the same value for Z that the machinist could use to offset? You are right in that it is a small section of a sphere, but the program is repeating and correlating with a gage ball.

[Matthew D. Hoedeman]

but the Xact wouldn't show a tolerance for the Z, yes, it should show the Z and the Z deviation, but I've seen too many machinists confused by position dimensions since the axis don't have a tolerance, and the position tolerance would be double what the Z axis tolerance would be. "Well, the position tolerance is 0.010, and the Z is only off 0.008, so it must be good" kinda thing....

[JacobCheverie]

SingularitY I have attached a bit of my code (minus move points). Do you see a reason why shift in Z is not occurring?

$$ NO,
================================================== ==
------------------- DCC Alignment ------------------
================================================== ==

L2 =LABEL/
COMMENT/OPER,NO,FULL SCREEN=NO,AUTO-CONTINUE=NO,
ENSURE PROBE IS LOCATED AT A SAFE DISTANCE ABOVE THE PART
ENTERING DCC MODE

MODE/DCC

PLN2 =FEAT/PLANE,CARTESIAN,TRIANGLE
THEO/<-0.0089,0.0326,0>,<0,0,1>
ACTL/<-0.0086,0.0327,0.006>,<-0.0047119,0.0014105,0.9999879>
MEAS/PLANE,8
HIT/BASIC,NORMAL,<-0.4231,0.521,0>,<0,0,1>,<-0.4224,0.5209,0.0029>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.0079,0.703,0>,<0,0,1>,<-0.0078,0.703,0.0052>,USE THEO=YES
HIT/BASIC,NORMAL,<0.4034,0.4934,0>,<0,0,1>,<0.4034,0.4 935,0.0075>,USE THEO=YES
HIT/BASIC,NORMAL,<0.6051,-0.0248,0>,<0,0,1>,<0.6051,-0.0247,0.0089>,USE THEO=YES
HIT/BASIC,NORMAL,<0.3876,-0.4836,0>,<0,0,1>,<0.3879,-0.4835,0.0077>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.0237,-0.5509,0>,<0,0,1>,<-0.0232,-0.5508,0.0078>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.4152,-0.4282,0>,<0,0,1>,<-0.4148,-0.428,0.0044>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.5971,0.0306,0>,<0,0,1>,<-0.5968,0.031,0.0032>,USE THEO=YES
ENDMEAS/

AUTO_1 =ALIGNMENT/START,RECALL:MANUAL_2,LIST=YES
ALIGNMENT/LEVEL,ZPLUS,PLN2
ALIGNMENT/END

CIR3 =FEAT/CONTACT/CIRCLE/DEFAULT,CARTESIAN,OUT,LEAST_SQR
THEO/<0,0,-0.2185>,<0,0,1>,1.5,0
ACTL/<-0.0516,0.0031,-0.2185>,<0,0,1>,1.4985,0
TARG/<0,0,-0.2185>,<0,0,1>
START ANG=180,END ANG=360
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO

CIR4 =FEAT/CONTACT/CIRCLE/DEFAULT,CARTESIAN,OUT,LEAST_SQR
THEO/<0,3.631,-0.2185>,<0,0,1>,1.5,0
ACTL/<-0.0444,3.6348,-0.2185>,<0,0,1>,1.4989,0
TARG/<0,3.631,-0.2185>,<0,0,1>
START ANG=180,END ANG=360
ANGLE VEC=<1,0,0>
DIRECTION=CW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO

LIN2 =FEAT/LINE,CARTESIAN,UNBOUNDED,NO
THEO/<0,0,-0.2185>,<0,1,0>
ACTL/<-0.0516,0.0031,-0.2185>,<0.0019867,0.999998,0>
CONSTR/LINE,BF,3D,CIR3,CIR4,,
OUTLIER_REMOVAL/OFF,3
FILTER/OFF,WAVELENGTH=0

AUTO_2 =ALIGNMENT/START,RECALL:AUTO_1,LIST=YES
ALIGNMENT/ROTATE,YPLUS,TO,LIN2,ABOUT,ZPLUS
ALIGNMENT/TRANS,XAXIS,CIR3
ALIGNMENT/TRANS,YAXIS,CIR3
ALIGNMENT/TRANS,ZAXIS,PLN2
ALIGNMENT/END

$$ NO,
================================================== ==
------------------- DCC Features -------------------
================================================== ==

PLN3 =FEAT/PLANE,CARTESIAN,TRIANGLE
THEO/<0.0077,3.6303,0>,<0,0,1>
ACTL/<0.0076,3.6306,0.0049>,<-0.0011604,-0.0009198,0.9999989>
MEAS/PLANE,8
HIT/BASIC,NORMAL,<0.6066,3.6676,0>,<0,0,1>,<0.6067,3.6 681,0.0065>,USE THEO=YES
HIT/BASIC,NORMAL,<0.406,4.0373,0>,<0,0,1>,<0.406,4.037 6,0.0058>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.0189,4.2457,0>,<0,0,1>,<-0.0189,4.246,0.0054>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.4319,4.0609,0>,<0,0,1>,<-0.4319,4.0611,0.0046>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.5972,3.6794,0>,<0,0,1>,<-0.5971,3.6797,0.0038>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.3926,3.0974,0>,<0,0,1>,<-0.393,3.0981,0.0054>,USE THEO=YES
HIT/BASIC,NORMAL,<0.0283,3.0463,0>,<0,0,1>,<0.0282,3.0 466,0.0033>,USE THEO=YES
HIT/BASIC,NORMAL,<0.461,3.2075,0>,<0,0,1>,<0.461,3.207 8,0.0045>,USE THEO=YES
ENDMEAS/

TIP/T5-A0B-90, SHANKIJK=0.0016, 0.0004, 1, ANGLE=-89.423

PLN4 =FEAT/PLANE,CARTESIAN,TRIANGLE
THEO/<0.0267,3.0574,-0.437>,<0,0,-1>
ACTL/<0.0268,3.0573,-0.4324>,<0.0007001,0.0017599,-0.9999982>
MEAS/PLANE,6
HIT/BASIC,NORMAL,<0.5299,3.4186,-0.437>,<0,0,-1>,<0.53,3.4185,-0.4314>,USE THEO=YES
HIT/BASIC,NORMAL,<0.4025,2.912,-0.437>,<0,0,-1>,<0.4027,2.9119,-0.4325>,USE THEO=YES
HIT/BASIC,NORMAL,<0.0378,2.6958,-0.437>,<0,0,-1>,<0.0379,2.6957,-0.4329>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.3476,2.8869,-0.437>,<0,0,-1>,<-0.3475,2.8869,-0.433>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.4846,3.3107,-0.437>,<0,0,-1>,<-0.4844,3.3107,-0.4323>,USE THEO=YES
HIT/BASIC,NORMAL,<0.0221,3.1204,-0.437>,<0,0,-1>,<0.0223,3.1203,-0.432>,USE THEO=YES
ENDMEAS/

PLN5 =FEAT/PLANE,CARTESIAN,TRIANGLE
THEO/<0.0025,0.672,-0.437>,<0,0,-1>
ACTL/<0.0026,0.6724,-0.4362>,<0.0006368,0.0001731,-0.9999998>
MEAS/PLANE,6
HIT/BASIC,NORMAL,<0.4586,0.4454,-0.437>,<0,0,-1>,<0.4588,0.4454,-0.436>,USE THEO=YES
HIT/BASIC,NORMAL,<0.3666,0.7995,-0.437>,<0,0,-1>,<0.3665,0.7992,-0.4362>,USE THEO=YES
HIT/BASIC,NORMAL,<0.0232,0.934,-0.437>,<0,0,-1>,<0.0233,0.9348,-0.4361>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.3733,0.8561,-0.437>,<0,0,-1>,<-0.3733,0.8569,-0.4364>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.4477,0.4631,-0.437>,<0,0,-1>,<-0.4477,0.464,-0.4368>,USE THEO=YES
HIT/BASIC,NORMAL,<-0.0122,0.5339,-0.437>,<0,0,-1>,<-0.0121,0.5339,-0.4358>,USE THEO=YES
ENDMEAS/

TIP/T5-A0B0, SHANKIJK=0.0016, 0.0004, 1, ANGLE=0.577
LOADPROBE/T1
TIP/T1-A0B0, SHANKIJK=0.0016, 0.0004, 1, ANGLE=0.577

DATUM_D =FEAT/CONTACT/SPHERE/DEFAULT,CARTESIAN,IN,LEAST_SQR
THEO/<0,3.631,-0.2185>,<0,0,1>,0.8779
ACTL/<-0.0063,3.6344,-0.2128>,<0,0,1>,0.8767
TARG/<0,3.631,-0.2185>,<0,0,1>
START ANGLE 1=0,END ANG 1=360
START ANGLE 2=-2,END ANG 2=-18
ANGLE VEC=<1,0,0>
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO

DATUM_B =FEAT/CONTACT/SPHERE/DEFAULT,CARTESIAN,IN,LEAST_SQR
THEO/<0,0,-0.2185>,<0,0,1>,0.8779
ACTL/<-0.0023,0.0026,-0.217>,<0,0,1>,0.8766
TARG/<0,0,-0.2185>,<0,0,1>
START ANGLE 1=0,END ANG 1=360
START ANGLE 2=2,END ANG 2=-18
ANGLE VEC=<1,0,0>
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO

$$ NO,
================================================== ==
------------------ Constructions -------------------
================================================== ==

DATUM_C =FEAT/WIDTH,CARTESIAN,OUT,LEAST_SQR,YES
THEO/<0.0158,3.3848,-0.2185>,<0,0,1>,0.437
MEAS/<0.0159,3.385,-0.2137>,<-0.001,-0.0011,1>,0.4367
CONSTR/WIDTH,BFRE,PLN4,PLN3

DATUM_A =FEAT/WIDTH,CARTESIAN,OUT,LEAST_SQR,YES
THEO/<-0.004,0.3066,-0.2185>,<0,0,1>,0.437
MEAS/<0.0257,0.3051,-0.2181>,<-0.0002,0,1>,0.4362
CONSTR/WIDTH,BFRE,PLN5,PLN2

$$ NO,
================================================== ==
---------------- Datum Definitions -----------------
================================================== ==

DATDEF/FEATURE=DATUM_A,A
DATDEF/FEATURE=DATUM_B,B
DATDEF/FEATURE=DATUM_C,C
DATDEF/FEATURE=DATUM_D,D

$$ NO,
================================================== ==
------------------- Measurements -------------------
================================================== ==

COMMENT/REPT,

.004 -B- Sphere Position

FCFLOC1 =POSITION : DATUM_B
FEATCTRLFRAME/SHOWNOMS=YES,SHOWPARAMS=YES,SHOWEXPANDED=YES
CADGRAPH=OFF,REPORTGRAPH=OFF,TEXT=OFF,MULT=10.00,A RROWDENSITY=100,OUTPUT=BOTH,UNITS=IN
COMPOSITE=NO,FIT TO DATUMS=YES,DEV PERPEN CENTERLINE=ON,OUTPUT ALIGNMENT=Datum Reference Frame
CUSTOMIZED DRF=NO
STANDARDTYPE=ASME_Y14_5
SIZE TOLERANCES/1,DIAMETER,0.8779,0.0003,0.0004
PRIMARY DIMENSION/POSITION,<D>,0.004,MMC,A,MMC,<size>,<dat>,<dat>
SECONDARY DIMENSION/<Dim>,<tol>,<MC>,<dat>,<dat>,<dat>
NOTE/FCFLOC1
FEATURES/DATUM_B,,
DATUM_B
Z:NOM=0
<NEW>
DATUMS
D1:NOM=0.437,+TOL=0.003,-TOL=0.003

[SingularitY]

Sorry man. I'm out of ideas at the moment. I agree with Matthew D. Hoedeman I would switch to legacy so you know it's being done properly.

[JacobCheverie]

SingularitY No problem, thanks for the advice. Unfortunately, Legacy does not support datum widths unless I am missing something. I am looking forward to 2020 R2, many of my headaches should be resolved by then.

[Matthew D. Hoedeman]

'datum width', you can calculate the bonus tolerance and then use that (divided by 2) as the +/- tolerance for the Z axis of the sphere.

Example:
'nominal width' 10
'measured width' 10.25
'width tolerance' +/-0.30
Bonus tolerance (M) is 0.55
0.55/2 = +/-0.275
Assign a variable to that and use the variable as the tolerance in the dimension.

about 47 ways to skin this cat.

[Matthew D. Hoedeman]

probably not it, BUT, add the level command to that AUTO_2 alignment. Pcdmis can and will do strange things if you don't have ALL DOF locked in a single alignment.

[JacobCheverie]

No dice, thanks for the idea though.

[Matthew D. Hoedeman]

have you tried a MID-PLANE instead of a WIDTH feature? You would still need to do some coding for the bonus.

[JacobCheverie]

I could do that, I was hoping to avoid that method so that I could stick with the default Xact report style. If parts start failing for this, then I may have to reconsider.