Measure tool not working correctly

[Noah_Katz]

Public link: https://a360.co/3Ktxxxj

This has been happening for the last couple of weeks in all of my design files.

The axes don't match the global coordinate system's, which isn't too bad because you can tell by inspection what they're supposed to be.

But a lot of the time the coordinate axes aren't even parallel to the global's.

[HughesTooling]

I think this is because measure is using the coordinate system of the component point first selected. You'll see it matches the spring post upper coordinate system. If you select the points in the other order it will match the belt's coordinate system. Also you should try fixing the errors in the timeline. Try and avoid using capture positions as well, if you roll the timeline back to just after a capture position and add a joint you can delete the capture position.

[Noah_Katz]

That seems very plausible, but if that's the case I don't see how it wouldn't have happened before now.

In the years I've been using F360 I've measured thousands of times between hundreds of parts that have been rotated and joined in all different orientations, and it never happened before.

[HughesTooling]

@Noah_Katz  Not sure if I'd notice, took me a while to see the colours didn't match the world coordinate system in your example. Just testing I rolled the timeline back in your file to where a component is slightly rotated and again the measure tool matches the coordinate system for the component who's point was selected first. @jeff_strater Is this something new? Actually quite useful, not noticed if it worked like this in the past.

[Noah_Katz]

Yes, it would be useful if there was a toggle, like via MB3, for global or component coordinate system.

[jeff_strater]

@HughesTooling - yes, this is intentional, and as far as I know, nothing new.  The owning component's coordinate system is used because if both points are on the same component, that is almost certainly what you want.  If both points are on different components, I suppose it is arguable that the first component's coordinate system makes no more sense than would the world coordinate system (or even the "least common parent" owning component).

I agree, @Noah_Katz - a toggle would be nice, but it's pretty low priority at the moment.

[Noah_Katz]

@Anonymous Strater

The behavior does now seem consistent with what you've described, but I still find it hard to believe it took years for it to manifest, as I use Measure heavily.

>  If both points are on different components, I suppose it is arguable that the first component's coordinate system makes no more sense than would the world coordinate system (or even the "least common parent" owning component).

I think it could be more strongly argued that if both components are rotated wrt global, the results should be in global, i.e. subassy parts wrt to a machine frame.