scanning direction and speed

[[An...]]

fairly new to calypso, I have a dmis background. trying to learn as much as I can in a short amount of time, much of the info I'm given is through other operators/programmers. I was told that when scanning (we have a vast and a vast xt head) that the direction matters. for reference most if not all of our parts are checked in vises sitting horizontally and can vary from 1 to 8 feet(think long tubes). on most diameter or cylinder scans we only scan 180 degrees starting from the top point in z down 90 degrees on either side with a probe in z-. I asked why we didn't use a probe probe in x-/+ and scan 180 on both sides of a part and was told that when scanning, pushing or pulling matters as far as when the stylus is in contact with the part and that we could not trust data if taking scans in this way. I was also told lowering the speed during a run will effect the data when scanning and could also be a problem. was hoping I could have some light shed on this as I haven't been able to pull any literature on this topic. thank you in advance.

[[Mi...]]

If all your probes are well calibrated you can indeed use multiple probes to more completely measure a circle, however this will virtually always throw in a tiny amount of error. The error gets compounded when calibration comes into question. You'll see it on a roundness scan.

Here is an extreme example, a circle measured with 3 probes on the same stylus system. After calibration the roundness dropped for 0.015.

So absolutely, completely possible, but be aware of the potential issues that I suspect people have ran into that makes them afraid to do it again.

As for direction and speed, we break all the "rules" there, there is a lot of application specific answers to that.

Roundness.PNG

[[An...]]

this all makes sense to me, as far as breaking the "rules" is there a good place to look for general guidelines or does it all come down to "learning by exposure" I have access to most zeiss training books/cookbooks. On a second look our general strategy for a cylinder lying horizontal from y- to y+ is to use a probe in z- and start in either x+/- and take 2 90 degree scans up to the z+ direction (one on either side) and of course repeat for the length of the cylinder. (I was backwards in my initial assessment). I'm more interested in scanning direction and how that matters over speed as most speed settings can be found in the cookbook I think. we do have a a mounted rotary facing y- but many of our parts are too heavy and I think I'm a ways away from adding that into my programming strategies.

[[Mi...]]

I've never known scanning direction to make a difference. I scan clockwise and counterclockwise based on program efficiency.

In your case it sounds like you could get the complete diameter with a +X facing and a -X Facing probe... each one scanning a little over 180° of the cylinder.

[[Ri...]]

Please sign in to view this quote.

In a time long, long ago....
I was told that Zeiss formulas were designed for probe travel in the direction that it automatically travels.
Circles and Cylinders are CCW.
Planes are CW.
I believe Cones are CCW.

From that, my OCD brain decided that I program in the same direction as a feature is automatically produced.
I have not tested this, I've been doing this the same way for about 18 years.

[[Mi...]]

Please sign in to view this quote.

I was told the same thing only 5 years ago, The Cookbook and I think Basic training book both say CCW for circles so there must be a reason, or maybe their was once upon a time and now it no longer truly matters. But I've experimented and found no appreciable difference in results, I have to program for cycle time, so one feature starts where the previous one ends. In the case of partial circles around a shaft that means alternating CW and CCW circles for efficiency