What is the different between selecting datums from Datum Ref. Frame vs Alignment

[[Me...]]

Can anyone tell me if there is any different when I report true position by selecting every single datum vs selecting an alignment which is created using those same datums. I have tried both way and they give out different result. Please see attached pics. Thanks in advance.

TP2.pngTP1.pngALIGNMENT ABD.png

[[Aa...]]

Technically, there is no difference, because when you select the datum features, it's creating an alignment behind-the-scenes.

However, there could be a difference depending on the specific settings you have for the datum features and the features in the alignment. (The defaults are sometimes different, depending on the settings in Calypso.) Are they being evaluated LSQ, OTE, or ? Are they being constrained to other features (preceding datum features)? You can see these specifics by right-clicking the icon for the feature either in the DRF section of the characteristic dialog or in the Alignment dialog.

[[Da...]]

Why is the first examples nominal position different from the second? It also looks like you offset -D- in one of the examples.

[[Aa...]]

It looks like rotation about Z was applied in the first example...

[[Me...]]

Aaron , The evaluation method and constrain settings are all same.

Dave That is my question. I have been told by Zeiss instructor that i can either select datum features separately or create an alignment and it works most of the time. I noticed that the alignment in the first example rotate to datum D (slot) on -Y direction, While the alignment in example 2 is same as base alignment. Should the true position be the same if they are calculate from the same origin ?

[[Me...]]

Please sign in to view this quote.

Picture 3 is how i create alignment ABD.

[[De...]]

Did you perhaps use the special button on your true position characteristic to change rotation in Z and not on the alignment?

Or did you perhaps have the rotational amount added to the alignment and to special in the true position characteristic?

From the pictures it is apparent that in the True Position characteristic picture, the alignment is aligned with the part axis.

In the alignment true position it is not.

So either the Alignment does not include the rotation it should, or the true position characteristic includes it so it is being doubled or something.

Definitely something up with the rotation though is my thinking.

[[Me...]]

please make sure datum Constrained is same on both alignment.

[[Me...]]

please make sure datum constrained is same in both alignment.

[[Aa...]]

Very odd.
The issue seems to be the use of a "slot" feature for the tertiary datum. (I can't say I've ever used that feature type.) It seems that when a slot is used for planar rotation in an alignment, Calypso uses it the same way it would a circle--the position of the slot relative to the X/Y origin of the alignment is what sets the clocking. However, when it's used as the tertiary in a DRF, it seems Calypso doesn't take the clocking from it at all. (I think that's a bug.)

Try this: create another TP characteristic using a DRF of only A and B, and see if the results are any different from one with A, B, and D.

[[Me...]]

I try using another circle (next to datum D slot) for the tertiary datum and the results come out exactly agreed.

Also create TP characteristic using DRF of A and B, and another one using DRF of A, B and D

True position.png

[[Me...]]

the correct way is to break out the datum's and not to use the alignment. It boils down to the settings you have in your form datum tab under your system set up (see picture)
I always tell users to have these settings set up on your machine.

if you are using your alignment you are defaulting to LSQ. if you are breaking out the datums you are using Outer Tangential. along the lines of what Aaron mentioned.

outer tangential element.jpg

[[Aa...]]

The OP stated that the evaluation settings, including constraints were the same for the features in the alignment as for the features in the DRF. If that's the case, the alignment should still evaluate correctly. (And given that it's taking -D- into consideration while the DRF for some reason is not, I would say that the alignment is the preferable way to go.)

If you right-click on each of the feature icons in an alignment and set them each to Outer Tangential Element, and constrain the secondary and tertiary to their preceding datums (which you should be able to do with the button on the Constraint tab "Form Datum as per ISO 5459"), then the alignment should behave like the DRF. (Of course, the OP seems to have stumbled onto a situation where they behave differently because of a bug that's making the DRF not use the slot.)

Be aware, the "Ref. Calculation as per ISO 5459" checkbox in the Measurement Methods settings of Calypso only affects the constraints on DRFs that are newly created. Old DRFs won't change when you turn this setting on or off. Nor will having this setting off prevent you from adding the constraints to each DRF (and Alignment) as you create it.

[[Ke...]]

Please sign in to view this quote.

Agreed.
I have used this method a lot.
All you need to do is go into the Special function, and either Rotate by Angle, or Rotate by Distance, to adjust the "clocking"

[[Me...]]

I try Both evaluation method to LSQ and OTE with those feature constrain to datum A. The result change .0002 for all the TP characteristics.

I try to adjust the clocking by rotate it around Z axis and it didnt work

Rotate.png

[[Me...]]

i also verify the true position using height gage and the result came out close to cmm result with TP using alignment.

[[An...]]

_

[[Me...]]

Please sign in to view this quote.

Is this what you mean?

Slot.png

[[Jo...]]

The difference is in whether or not the ISO 5459 is turned on. Apparently yours is. When you select each datum feature in your characteristic Calypso ensures the datums are mutually perpendicular if ISO 5459 is checked on. It does not do this with the base alignment or any secondary alignments unless you actually go in and constrain the features appropriately, as described by others in this thread.