Measuring a small radius

[[Su...]]

I have a valve leaf .381mm thick. The edge radius is .095mm. Probe is 1.5mm. I need to measure the radius. I measured a 2D Curve on the edge radius and made a theoretical circle with the points to get radius. I lessened the probe pressure from 100 to 50(too low?). Is this correct measuring process? Is there  a better way?

Using a Contura Select with a Vast XT Gold.

[[Ma...]]

Keep slow scanning speed and play with density of points - then it's ok. 50nM is ok too.

Are you having any problems with stability of results?

[[Cl...]]

If it is a small circle segment <180°, then under the circle feature evaluation, constrain the two axis (Which ever work-plane the feature is in). 

In addition, I would use single points instead of a curve. If the thin material is wavey, I would probe single points on the top surface at the same coordinates as the radial points and use a formula to get the correct Z value to help compensate for variations in Z due to the waviness.

Edited June 4

[[Su...]]

Working without a print, guessing at some feature angles and radius at profile area. Since the area is so small not sure if scan is perp to the part. Scan speed at 1.0 tends to jump a bit but deleting those flyers leaves me with 16 pts on the curve to construct a circle. Nitrile probe and not ruby, smallest probe I have. Yes, <180 coming in at 88.1deg.  Not sure if my result is 100%.

 

[[Ma...]]

I would invert direction of scan - from 30. point to 1. point.

I would also do inital bestfit to have this curve measured as close to the real part as possible.

[[Su...]]

Thank You to Martin Jansky and Clarke Gilbert for your responses! Mission accomplished!