Self-centering point suggestions

[[To...]]

Looking for suggestions for settings on a self-centering point.  In the pic below, view is from -Y)  I am using a 0.8 stylus to measure the depth of a step that is 0.5 wide. Though this will be done with VAST Gold head, sometimes when using a star probe on an XXT, there is a tendency for the probe to glance of the step and slide past the step.  I'm thinking a self-centering point with some force towards the side wall, along with force in the probing direction might help.  I am also taking a point on the side wall to help control the location of the point on the step, so I can get as close as possible.  I am also considering lowering the probing dynamic.  How would you guys approach this?

[[Cl...]]

Maybe I'm not quite understanding exactly what you are trying to do, but why not measure a line on each surface and intersect them for a point?

[[To...]]

Please sign in to view this username.

 I am measuring the depth of the step where the tip of the probe is at and want to make sure the probe doesn't slip off the step.

Edited July 25

[[Cl...]]

OK. I didn't see that at first. I thought it was just the corner of some square geometry.

Edited July 25

[[Cl...]]

On that wall parallel to the shaft probe a point. Create a secondary alignment using that point (XYZ).

Place a nominal point feature on the other surface recalling that secondary alignment. insert a formula to put the point at 0.45 in Z.

 

Edited July 25

[[Cl...]]

I guess you don't need a formula. Just 0.45 in Z from the secondary alignment.

[[Ke...]]

Exactly as you said: I have often needed to execute a self-centering point with some force directed at the side wall. I would target 45 degrees, use a midpoint evaluation, and use a formula to calculate the point of contact.

[[To...]]

Please sign in to view this username.

 Seems like a touch point would still calculate to the tip, but I will try both and compare.

[[Ni...]]

I'd personally just try a single point with the feature aligned on the nearest surfaces with the probing dynamic turned down to 12.5%

[[Ri...]]

In the past, I would probe a point on the side wall, use that for my location control from the side wall (PCM) to probe the point you require.

Now it will always be X amount from the side wall ensuring your contact is good for the point you require.

[[Ke...]]

Although I am loathe to do any trial and error with a 0.8mm, it may be worth experimenting with the clamping of the machine axis during self centering, or using the "own selection" to set the directional forces. Adding a bit of downward force will likely help keep the probe from slipping off. My screenshot is from a +Z point in an empty program, so the Fz is simply -200 by default

[[To...]]

I had trouble with using "Own Selection".  No matter which values I used, it would slide off the step.  I ended up using a touch point but modified the vector to come at an angle less than 45° and used "Probe in Normal Direction".   Haven't confirmed the results yet.

[[Ke...]]

[[Ri...]]

If you are going to probe in the direction, I would personally feel safer using mid-point, and just offsetting the probe radius manually. I'd be nervous of incorrect vector compensation. 

[[To...]]

Please sign in to view this username.

  I did both and got the same exact result.  Touch point reports to closest primary axis, which is X, so vector doesn't matter, as long as more X than Z.

Checked depth in Z direction with .5 probe and result was within .0002".  I'm going home and taking my ball with me. 

Edited Tuesday at 07:44 PM

[[Ma...]]

Tom, if possible, put a known gage pin in the hole touch the top and use a formula to subtract the difference using Result Element.

[[To...]]

Please sign in to view this username.

 Not a hole. But that does remind me that I have had relatively good results using Step Points.

Edited yesterday at 01:17 PM