External Thread runout

[[Ju...]]

Hello,

I am measuring a threaded rod and checking the runout of the thread to the non-threaded cylinder. I have it setup 2 different ways- one to helix scan the threads to the pitch diameter (M8x1.25) and one to scan 8 vertical lines around the threads. Which would be more accurate? I have filtering turned on for both, using LSQ. Also, using a 5mm stylus tip.

Thank You

[[Jo...]]

I use a small probe to do short line scan across thread. Take the max or min coordinate, depending on which quadrant you are working on. That will be the trough centered in the pitch. Use a formula in your helix strategy to recall that coordinate + 1/2 pitch and it will scan on the peak of thread every time. It works very well for runout.

[[Ch...]]

I recommend a self-center helical scan. Flood it with points
then filter and OLE the crap out of it.

[[Ri...]]

I have been using Line scans on ID and or OD threads for about 9 years now.
In order to use line scans and not have errant A1 and or A2 results I define the feature as a circle.
Once the circle is opened change it to a Cylinder.
Set generic Filter and Outlier options (Defaults fro Calypso)
(Some of the settings)
In threads of 24 pitch or finer these are my generic settings.
Speed 0.0906
Step Width 0.0039 (Dependent on Target height)
No of Lines 4
Angle Range 360 (I have used less than 360 on many occasions)

As for Run out, I create a circle at the entry and at the greatest depth I can (Requires two run out characteristics). If either fails, it fails. It is rare that we have a run out issue; most likely due to our machining process for tapping and or thread milling.

I have attached a couple screen shots of what I have been doing for about 9 years.
If I can find my old Excel files (On an old PC) I have several years of Hole size data and Location data with Cpk's in the 14.x range for diameters. Location data was tested against Tru-Pos locators and held a respectable 6.x Cpk.

I have yet to have a customer reject this method or our part data.

Calypso Thread screan shot 1.jpgCalypso General Thread Scanning Guidelines 1.jpg

[[Ri...]]

I am rather fond of the helical self-centering scans. If you are going to scan the linear line scans, change your outlier elimination to be more aggressive (usually 1 - with an iteration loop of 3-5) on inside the work piece if working with an internal thread (I assume that it's the opposite on an external thread, but I don't measure much of those).