circle

[AuRules]

How does two (2) half circle under the mean = a larger diameter (see below) I am using 2 tip angle to get the 4 points per side. Is it in the way i'm creating a circle?


MOVE/POINT,NORMAL,<-1.243,0,2.5>
CIR23 =FEAT/CONTACT/CIRCLE/DEFAULT,CARTESIAN,IN,LEAST_SQR
THEO/<-1.243,0,1.04>,<0,0,1>,0.5005
ACTL/<-1.2416,-0.0025,1.04>,<0,0,1>,0.4996
TARG/<-1.243,0,1.04>,<0,0,1>
START ANG=55,END ANG=305
ANGLE VEC=<1,0,0>
DIRECTION=CW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
MOVE/POINT,NORMAL,<-1.243,0,2.5>
TIP/T1A7.5B180, SHANKIJK=0.1367, -0.0007, 0.9906, ANGLE=-90.2496
MOVE/POINT,NORMAL,<-1.243,0,2.5>
CIR24 =FEAT/CONTACT/CIRCLE/DEFAULT,CARTESIAN,IN,LEAST_SQR
THEO/<-1.243,0,1.04>,<0,0,1>,0.5005
ACTL/<-1.2424,-0.0025,1.04>,<0,0,1>,0.5002
TARG/<-1.243,0,1.04>,<0,0,1>
START ANG=125,END ANG=235
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
CIR25 =FEAT/CIRCLE,CARTESIAN,IN,LEAST_SQR,NO
THEO/<-1.243,0,1.04>,<0,0,1>,0.5005
ACTL/<-1.2427,-0.0029,1.04>,<0.0000904,-0.0000008,1>,0.5014
CONSTR/CIRCLE,BF,3D,CIR23.HIT[1..CIR23.NUMHITS],CIR24.HIT[1..CIR24.NUMHITS],,
OUTLIER_REMOVAL/OFF,3
FILTER/OFF,UPR=0
MOVE/POINT,NORMAL,<-1.243,0,2.5>

[Matthew D. Hoedeman]

maybe try bestfit recomp?

[DungT]

cir23 X = -1.2416
cir24 X = -1.2424

[ajlee7]

When you go to your autocircle settings, there's a button that says "To Points" (It'll be in the same area where you select circular movement).
If you select this, then both of your half circles will be created as a series of points instead of a circle.
Then, you can construct a best fit circle out of those points.

Hopefully that works for you

[JacobCheverie]

You are performing a least squares fit and each half circle has a different center point. The math must just add up. If I have a Ø.5 half circle at X=0 and an opposing Ø.5 half circle at X=100, the least squares circle will absolutely be larger than Ø.5.

[AuRules]

I will play with it, thanks for the suggestions. same circle, 2 different tip angles. only Heaven knows.

[louisd]

You've got 110° of arc segment, so that should be enough to get a robust radius. There's an old write-up in hexagon's knowledgebase on how to measure the diameter/radius of a partial circle.
In short, measure the arc segment with a FIXED_RAD circle (set to your radius nominal).
align/origin to that circle.
create auto-points change your coord method from Cartesian, to Polar. the polar radius of your points is equal to your size.

I would apply this same process to your interrupted diameter. Measure it. center to it, then measure it again. output your second set of data points.

[JEFMAN]

First, I would check the calibration (tips diameters and std dev).
I would also construct a 2D circle.
BFRE should give better results, only if tip diameters are very close one of the other.