New user with a question about 'thread-following' in Calypso.

[[Da...]]

Long-time PC-DMIS user, new to Calypso (and getting frustrated...); We've got a new Spectrum III (II maybe??), with Calypso 2024. We make certain types of electrical assemblies which have a LOT of PEMS. With DMIS, it was easy to enter the pitch and diameter of whatever PEM I'm measuring; the only thing slightly difficult was remembering that non-metric pitches are inverse, but that's about it.

TODAY, I'm measuring a PEM for the first time on my brand-spanking-new Zeiss, and I can't figure out how to do it.  I've found 'helix measuring' on cylinders, but nothing is obvious on how to follow say a .5mm, a .7mm, or a 1.0mm pitch. Can anybody help me?

I'm sure I'll have more issues in the days  to come.

[[Ro...]]

Try doing the thread as a circle.  In there you will find helix path.  Towards the bottom, you can enter the number of turns you want to do and the slope, which is your pitch.

[[ni...]]

Hi,

for thread I mesure circle with two trajectory of 4 points, the second one with an offset half of the pitch (the M10*1.5 is 0.75 hight), and a Ø stylus at list 2* the pitch( 3* is better)

simple and work very fine if the pre-hold is straight in line.

Edited November 26, 2024

[[Ke...]]

Please sign in to view this quote.

As Robert mentioned: Slope = Pitch. Note that under "Special Settings" tab is the "Self-Centering" option; "Probe in Normal Direction" should be the appropriate selection.

[[Mi...]]

Spectrum is XXT. XXT won't reliably self center on a thread.... tried it many times, the repeatability is unacceptable.

What will work, and is one of the Zeiss recommended methods that was derived from a lot of testing.... and I've confirmed it to work well in our own process is:

 

If only a circle is required, scan a helical path that matches the thread pitch. 1.5 - 2 Revolutions.

If a Cylinder is required, scan multiple helical paths at 2-3 heights that match the thread pitch... 2 Revolutions each.

With no self centering in either method. The argument is, the entire thread was cut together, so it matters very little where on the tread you're touching as if you match the thread pitch the probe will be touching the same place all the way up.

If you do what I suggested vs the self centering version of the method on an XT machine, you're results will be virtually identical, because XT can self center a thread reliably... if you try it on an XXT machine, the self centering version will be all over the place. Yes, being the "Most" correct is important, but sometimes the more stable method is importanter.....

 

[[To...]]

Please sign in to view this quote.

Quite often, the minor diameter on an ID thread and the major diameter on an OD thread are cut separately.  One exception is when threads are created using a forming process, i.e. roll form, cold form.  In those case, the resultant minor or major diameters may not be fully or consistently developed.  For turned threads, there are some threading inserts that cut the entire thread profile.  For tapped holes, we hope the tap follows the hole. 

Edited November 26, 2024

[[Mi...]]

True but, we are talking about threaded holes in this thread here, using inserts in this specific case - where any form error in the thread is going to be very minimal. But it works equally well for an external thread.

And, that doesn't change the fact that XXT can't reliably do it with self centering, so the slight error from the threads being cut with other methods is less than what XXT will produce trying to self center and attempt to follow the pitch... 

And, unless you're talking very large threads with very small rubies, you're not scanning on the pitch diameter regardless, so even a self centered ruby is going to mostly ride on the Minor or Major depending on Internal or External.... so it's a moot point that they may not get cut at the same time. You're still better off making sure it at least rides on the same thing all the way up.

My answer come from Ryan Stauffer... It's fast, it's reliable and most importantly it's repeatable for all CMM types, even when our customer tries to compare our results to their Mitutoyo CMM..

I would post the PDFs from that class... but they say all over them I'm not allowed to. It's covered in his very informative Continuing Education Series 2 for anybody curious.

 

 

[[Da...]]

Thanks for all the answers. As an engineer, it never once occurred to me that slope and pitch could ever be considered the same thing...

[[Da...]]

Please sign in to view this quote.

Yeah, I'm beginning to see that the XXT head ain't nearly as good as that Renishaw indexer we used to have. In fact, my Spectrum seems to be a 'budget' CMM.

As for thread following, yes, that's the whole point; it doesn't matter whether you are on the peak, the side, or somewhere in-between, it's just that you will follow that same elevation of the thread all the way around, making a more reliable statistical center to measure to.

 

And, as for Mitutoyo CMM's; Wow, I haven't used one of THOSE things in 20 years! They were BANNED in the states back around the early 00's because they sold a couple to Gadhafi, and I don't think they ever re-entered our CMM market.... 

Edited November 26, 2024

[[Mi...]]

Mitutoyo is slowly coming back, but I'm in Canada so maybe it's just here they've returned.. they have Renishaw equipped machines and the machines themselves are "fine", the software is much more limited, it's a lot like DMIS 25 years ago.... A couple times now I've had to play correlation games with Mitutoyo CMMs and that's painful.

I recently turned down a job offer because I didn't want to deal with Mitutoyo CMMs...

Don't hate the XXT too much, and don't try to program like that Renishaw you're used to, Program Zeiss like Zeiss and you'll be fine. If programmed well you will probably find better repeatability, and correlation to other measurement types with the Zeiss machine than with other CMMs.

Please sign in to view this quote.

 

[[Da...]]

OK, now I know how to 'thread follow' in Calypso, but I still have one dimension on this nightmare of a part to measure. It's an M6 PEM, mounted on a formed tab, REALLY close to the bend, and inaccessible straight-on in the Y+ stylus position. So, as a result, I need to measure it from the Z- stylus position. As of right now, I've done a janky measurement with 3 separate arcs of 160° with 4 points each on the side of the PEM, but was wondering if there is some way to get thread-following on multiple arcs of less than 180°?

[[To...]]

Please sign in to view this quote.

Understood. My comments were related to threaded features in general.