Using alignment or Datum Reference inside characteristic

[[To...]]

Well is quite a long name to my topic.

Is there any difference in constructing a new Alignment or using DRF inside a characteristic?

[[Ja...]]

It is my belief that there should be no difference as long as you set your alignment up exactly the way that the DRF will constrain your degrees of freedom and you evaluate your datums the same way (LSQ, Max/Min Inscribed/Circumscribed, etc...)

Example: If your primary datum can constrain X-translation and you input it as a DRF, it will do just that. If for some reason you have an alignment with another feature constraining X-translation and you use that, it'll not report the same result (in X).

[[SH...]]

Please sign in to view this quote.

Secondary alignment is used when your drawing align your features from more than one origin , otherwise no need of secondary alignment.

As a smart programmer you should use DRF for your true position, profile tolerance etc characteristic.

[[Jo...]]

Please sign in to view this quote.

Can you explain the purpose of doing this ? Our instructor warned us against using alignment and Datums as if they were equivalent. Depending on your settings, the values at the feature level will not be the same as at the Characteristic level. Filters, evaluations, and hierarchy will influence that.

[[Ri...]]

The biggest difference is following the Can/May/Must rule. A lot of times people don't build their secondary alignments following the correct building of their DRF, which will lead to the differences in using one or the other.

There are also some rare cases where the DRF isn't built correctly because Calypso tends to not like using points as the tertiary, but the secondary alignment will be built correctly.

[[Ke...]]

"Back in the day" I was told that secondary alignments are for navigational purposes only, and to build the datum references in the characteristic. For the most part, that is my rule of thumb; however, I like to compare the two methods, as a sanity check.

[[Ma...]]

Secondary alignments are most useful for navigation and calculating the geometry of a feature to a different coordinate system than it was measured to. There are some differences between the way alignments and datums are calculated. Most notably, in the case where more than one element with an axis is used for a DRF the secondaryand tertiary datums are set to zero depending on the behavior specified in the config setting , when an alignment is calculated, the secondary and tertiary datums are set to zero where their axes intersect with the other axes.

[[Ri...]]

Please sign in to view this quote.

This

[[To...]]

Please sign in to view this quote.

No I can't. That is way I am asking 🙂

I am just wondering I have to options to basicly doing the same. I prefer the DRF, but sometimes I have tried using an alignment and I am getting small deviations from DRF to the alignment. That is way I were wondering if there are some basic settings which always apply to one or the other.

Thank you for all replys.

[[Jo...]]

Please sign in to view this quote.

There is a global setting for Datum measurement methods. Choices are LSQ or Outer Tangential, and you have the option of selecting ISO 5459 method of Datum relationship. You'll have to get someone else to expand on that.

[[Ma...]]

Please sign in to view this quote.

Here is a link to explain Form Datum, what John is talking about. http://www.mbccmm.com/cmmq/index.php/tr ... form-datum

Try an experiment yourself. Turn on Additional Positional result and create a TP with the measured datum. Now create another TP with a constructed datum and you will see a slight difference. When you construct a datum within TP it releases the degrees of freedom and reestablishes a DRF. This is the only possible way if the true position callout is to 1 or 2 datums.

DRFs can be different to alignments, such as datum order, opening a degree (either a rotational or translational) of freedom.

Now this is not say you must always build a DRF inside TP but I would highly recommend it when possible. I agree with some alignment complexity it may not be possible.

Mark