Mimicking pcdmis xact alignment in legacy mode

[BIGWIG7]

Is there a way to mimic how PCDMIS calculates and creates its alignment in XACT measure but do it as a Legacy mode alignment?

I have a complicated alignment and when I label the datums and build the XACT measurement DRF and then attempt to copy it exactly in Legacy mode I get a slightly different result for my profiles.
I want to know how PCDMIS is aligning the part in Xact so I can replicate it in Legacy.

Thoughts?

[NinjaBadger]

Does the datum system lock all 6dof without any MMB?

[VinniUSMC]

What type of features are your datums? More info please. Spell it all out.

[AndersI]

XactMeasure uses the mantra CAN - MAY - MUST - if a datum CAN lock a dof and MAY do it (not already locked by previous datums) then it MUST..

Additionally, secondary and tertiary datums are re-calculated internally with the additional constraint of being at right angles to the primary - this may be where Xact and your 'normal' alignment differ (different origins).

With MMB on the drf, it's getting difficult to replicate with the ALIGN command, as the MMB fit depends on all the dimensioned features with the same drf. There is no way (that I know of) to do a bestfit of all the concerned features, constrained by the freedom of the drf…

With PC-DMIS 2019 R2, you have new feature constructions for constrained secondary and tertiary features, but in earlier versions it is a bit more convoluted. Matthew D. Hoedeman (and maybe others, too) has explained well (somewhere here on the forums) how to do it with cylinders, I have done some experiments for planes, but I don't know of a single place 'cook book' on how to do it for every kind of feature.

[BIGWIG7]

2 planes not perpendicular to each other and a cylinder perpendicular to the secondary datum.

A= primary plane
B= plane not perpendicular to A
C= cylinder perpendicular to B

I guess what i'm looking for is how XACT is rotating the part using B plane

[AndersI]

I don't know how Xact handles that case - maybe it does re-calculate B with the constraint of having the correct theoretical angle to A, but I haven't seen that explicitly documented.

[Sora5]

When you have your cursor over an Xact alignment, the cad screen should show the location of the simulated origin based on your FCF. Just derive where that's at using your features and you can construct an alignment that puts you in the same spot. As long as there is no material conditions in your Xact FCF, your results should be similar using legacy.

[BIGWIG7]

When you have your cursor over an Xact alignment, the cad screen should show the location of the simulated origin based on your FCF. Just derive where that's at using your features and you can construct an alignment that puts you in the same spot. As long as there is no material conditions in your Xact FCF, your results should be similar using legacy.

yes I know that. what i'm saying is that when I try to replicate it as best I can how do I know I am in the same exact spot?

reporting profile in XACT gives me a different answer than using LEGACY even though it LOOKS like the alignments are the same.

HENCE, the request for the ability to use XACT's alignment and recall that as a conventional Legacy alignment.

It's not as easy as just looking at the trihedrons and seeing they are in the same location.

[JacobCheverie]

XACT is performing a mathematical simulation of datum shift when you have MMB. That means the software is algorithmically moving your part around your DRF ever so slightly (dependent upon how much the datums are deviating from their datum feature simulator at MMB) to minimize the profile result. I can imagine these algorithms are computationally difficult. What are your datum shift values from your XACT report?

Also, keep in mind that the method proposed by Sora5 is sufficient if you note the caveat "As long as there is no material conditions in your XACT FCF" (on datums, specifically).

[AndersI]

I think there was a thread, in the Code section, on how to get that coordinate system. I don't remember if it came to any conclusion...
(time to leave for home now, but I may search for it tomorrow)

[BIGWIG7]

but you do admit, it is not an easy task.

[Sora5]

Do a sample position dimension using Xact FCF and then do it again except use Current Alignment. If the axis nom/meas are the same, your alignment is replicated correctly.

[Matthew D. Hoedeman]

(1) Make your alignment to the datum features you will use in Xact
(2) Dimension what you are dimensioning using Xact (and defining your datums from those features)
(3) Dimension what you are dimensioning using Xact (BUT tell it to use current alignment instead)
(4) Compare the results
(5) Dimension what you are dimensioning using Legacy
(6) Compare the Legacy to the current alignment Xact.

[JacobCheverie]

I thought that switching between DRF and Current Alignment only changed the "look" of the report in that it may make more sense, but the actual calculation will be identical. BIGWIG7 has yet to respond to whether MMB is applied or not.

[Sora5]

The calculation is identical, hence your result does not change. What does change is the axis nominal values as it reference back to the FCF or current alignment. If noms for both are the same, that means your current alignment is the same as what Xact simulates.

[BIGWIG7]

MMB does not apply.

[BIGWIG7]

(1) Make your alignment to the datum features you will use in Xact
(2) Dimension what you are dimensioning using Xact (and defining your datums from those features)
(3) Dimension what you are dimensioning using Xact (BUT tell it to use current alignment instead)
(4) Compare the results
(5) Dimension what you are dimensioning using Legacy
(6) Compare the Legacy to the current alignment Xact.

I can and was able to do this.
sounds like there isn't a shortcut to be able to RECALL/COMPARE the alignment that XACT uses.

thanks Matt.