How to find center of a Spline Pitch

[[Ri...]]

I have a spline 15 Teeth, 20/40 Pitch and 30° PR. Ang.
It is not clocked to the part datums.
How can I find the center of the Pitch to clock it for measurement of Major and Minor diameters?

[[Ma...]]

Maybe this will help.

[[Ri...]]

Thank you, this looks like what I need.

Edit:
This is good if the Spline is timed correctly, in my case, the spline is not timed to any feature on the part.
The parts are hand loaded and randomly set on the arbor, there has been no need for alignment until now.
Hopefully in the future it will be timed.

[[Da...]]

Eye ball it ! or make a mark with a sharpy. You only have to get it close enough for the self centering point to get between the teeth... Ive done this by letting the probe get close, stopping the probe, adjusting the part, then letting it proceed.

[[Mi...]]

We find the rotation of random splines all the time - fully automatically, if I'm understanding that's your issue.

A Point, Self Centered, Mid point eval, 2 instance rotational pattern set to half the angle between 2 teeth. This ensures one of the points will self center in the tooth space.
This spline has 38 teeth, 360/38/2 = 4.73684
Minimum or Maximum Feature that takes the Min or Max that makes sense. -Z in my case on this part as it is loaded in the Z/Y plane and the one that Self Centers will have the lower Z value.

Use that Minimum feature to rotate your coordinate system.
Measure anything you need to measure for the spline on this rotated coordinate systems

[[Ri...]]

I'm glad to see other people use this method. Has worked well for me for many years with splines, as well as various other features (large threads as well).

Edit: The most common mistake will be that someone will forget to switch to Mid-Point Evaluation, so the values will be perfectly located at nominal - lol.

[[Mi...]]

Please sign in to view this quote.

I've spent more time than I'd like to admit being frustrated at this method not working, only to look at it the next day and realize I had the points set to touch point....

[[Ri...]]

Please sign in to view this quote.

The part uses a diameter Datum as a -Z alignment, it is set and can't be moved.

[[Ri...]]

Please sign in to view this quote.

So far, this method is working.
Thank you.

[[Mi...]]

Please sign in to view this quote.

If you do this on an XXT machine you may have to lock 1 axis parallel to the axis of your spline, XXT for whatever reason tends to want to slide along the teeth.

[[Ma...]]

I can offer another method to find the center of a tooth space, perhaps more quickly than the two point method. Just program one self-centering midpoint as described for the two point method, but before you run the part edit the vectors so the probing occurs at a small angle (10 or 15 degrees should do). For example a 10 degree angle from the +X-axis would have vectors i = .984808, j = .173648, k = 0.

The small change in probing direction is enough to cause the machine to slide off of the top land of a tooth if it happens to land on one the self-center in the next slot. If it happens to land within a slot it will self-center there.

This has worked well for systems with Vast Gold sensors. Obviously this would not work with a trigger probe and we have not tried it with other scanning probes.

[[Ri...]]

Please sign in to view this quote.

I am seeing this randomly on an XTR head. I can watch the stylus ride the flank.
My Spline is 24°, this has caused the stylus to fall into the groove on both attempts and conversely stay on top of the minor diameter. I will have to work on how to refine the timing.

[[Je...]]

Please sign in to view this quote.

Not possible with an XXT. Do not attempt this method. The XXT cannot self-center perpendicular to its sensor axis. Parallel to its axis requires per-application testing to ensure proper results are achieved.

You must instead use a 2d-line and scan over the spines to find the min/max coordinate of the feature. This must be done with a very slow scan and with very high point concentration while a larger stylus than you would typically use to self-center appropriately.

[[Ch...]]

Jeff,


Thanks for this info.. Do you have any Zeiss documentation on the XXT and self-centering ?

We just got our first DuraMax for the shop floor and may require some self centering for gear teeth and gap locations.

Thank you.

Chris

[[Mi...]]

Please sign in to view this quote.

Well... we do it. Every single day, thousands of times a day on 25 different XXT machines, with results that correlate within microns to XT Machines.

With gear carriers being our bread and butter, they very often have a Spline Minor Diameter as the secondary datum. In fact the program I used screenshots for runs almost exclusively on XXT and it Self Centers no problem. As long as we lock 1 Axis, Generally X due to how we load parts.

[[Ch...]]

Michael,

Thanks for the info, we are a gear shop as well. We just got out first XXT Duramax for shop floor.

Can you also confirm if you have any parts with a gear tooth or gap is a timing or alignment reference, e.g. a tertiary datum ?
And the results you get on XXT are close to what you get on XT ?

thank you !

-CR

[[Mi...]]

Please sign in to view this quote.

Unfortunately, or maybe fortunately for us, the only parts we have where the teeth are timed, the spline teeth have straight sides and are large enough to simply scan with a 2-3mm ruby.

But, we have had to do extensive comparisons between XXT and XT, including verifying CMM results with a customer in Germany because they were (rightly) mad at us for a quality issue... and our CMMs agreed completely with theirs. This was on one of the oldest parts we make that relies on the CMM self-centering on a part that dubiously uses the pitch diameter as a secondary datum of an involute spline.

We also frequently have to verify results from one CMM to another and as long as everything is clean and calibrated, XXT and XT both give us the same results even in cases where we are self centering in ways that apparently aren't possible.

The caveat is, we run the exact same program on both, meaning the program is setup for XXT. We don't take advantage of the higher potential speeds of XT because then I need to make different programs for different CMMs. 33 CMMs, 2 programmers - there's no time for that.

[[Ch...]]

Michael,

Ok sounds great. Thanks for the helpful info.

I thought I have used it once or twice before on XXT and have also seen similar results to VAST and parts were conforming.

We will likely do some comparison tests on our new DuraMax - with the understanding it is inherently less accurate than a Prismo etc.

Cheers !

[[Je...]]

Please sign in to view this quote.

Chris - See attached.
See page 96 (5-18)

While the XXT self-centering option may function under certain versions of CALYPSO, it may not with others and is not officially supported as a sound, repeatable measuring process. Use at your own risk and only after adequate testing.


I would advise EVERYONE read this as it will clarify dozens of questions that appear here quite frequently.


This is publicly accessible on the PORTAL - DOWNLOAD CENTER - DOCUMENTATION - SENSORS
https://portal.zeiss.com/download-center/documents