Thread measure

[[Br...]]

Hello,

I have a lot discussions with colleges of mine on how to measre position of a thread. One says cilindrical with the helical pitch. The other just a circle with the helical pitch. And another one just a circle with 7 points on one height. No pitch. What are your opinions? Thanks for the answers in advance!

[[SH...]]

If you are measuring with CMM , you can measure a cylinder and change the evaluation to maximum inscribed element. Try to measure three or four segment with more points.

[[Ma...]]

I normally measure a circle with 5 points and use the helical pitch. For most of the position tolerances this method is sufficient.
In my opinion the most important thing is to use the helical pitch to get the "correct" position of the thread.

[[Ri...]]

Of all the applications that I've measured it's hard to say which is the "best" as that is subjective. A lot of it depends on the thread (how big it is, how good the form of the thread is, and also how tight the position tolerance is).

I've measured a lot of different threads, and while most of the time you can get by with the linear line scans, and/or the helix path traces, the most accurate way that I've found is with true-position plugs (not the flex plugs). They have been the most accurate, and most repeatable for me, but they are expensive, and time-consuming.

If you are doing linear line scans, my suggestion would be to crank the step width way up, go slow, and for your outlier elimination switch to a 1/3 outlier, 5 adjacent points, 3 iterations. This will allow you to use the LSQ evaluation instead of having to use the unstable MIC/MCC evaluation.

[[Th...]]

Owen Long uploaded a PDF covering this topic last year, I'll link to it here:
viewtopic.php?t=1471#p9562

Doesn't really provide a 'best' solution but it has some nice pictures for visualizing the results of the different strategies.

[[Br...]]

Thanks for all the answers till so far! I shall test them for shure.