assembly mate problem sw2006sp4.0

I'm wondering if anyone else is seing something like this.
I have a linkage mechanism (simple link members with hinge pins) where I
want to make assy configs based on distances from the end of the moving
linkage to one of the assembly reference planes. The last component on the
end of the linkage can be mouse dragged around with the "Move Component"
icon and the linkage moves as expected. If I try to create a distance
mate from a point on this same component back to a plane, I get a message
saying "Warning: This mate is over defining ....." as soon as I click the
distance icon. I can OK the warning message and then put a value into the
distance mate, and it sometimes gives a message "Plane and point are not the
correct distance apart. Actual distance is xxx, desired is yyyy" where yyyy
is the distance I want as a mate. On one occasion it applied the distance
without this message and got me back to the assembly with a bunch of
overdefined mates, mostly on the hinge pins.
This is a very simple assembly with no limit mates or anything like that.
Mate Diagnostics doesn't appear to be functional at all.

Is there a serious bug in the mates function again?
(I remember fighting with stuff like this in the last millenium}

Thanks,
Bill

 

SWX seems unreliable in respect of what you are trying to do. It frequently
says things are over-constrained when they are not - this applies to both
sketch relations and assembly mates.

One thing that may help (but you may be doing this already) is to ensure
that all the mates that you are using prior to adding the distance mate do
not try to control more degrees of freedom than necessary.
By that I mean that (for example) a face-face coincident mate removes 3
D.O.F. , but the system will let you constrain 2 cubic components with 3
face-face coincident mates, which in theory accounts for 9 DOF - but of
course we know that in the real world there are only 6 DOF.
In other words, it allows you to over-constrain the components so long as
the result is consistent.

Try where possible to use the 3-2-1 rule i.e. first mate removes 3 DOF, 2nd
mate removes 2 DOF and 3rd mate removes the final DOF.
In practice, this tends to mean using fewer face-face coincident
constraints, and more face/plane-point coincident, and face-face parallel
constraints.

It'll probably still fail, but at least you'll have the satisfaction of
being able to swear about the shit software!

John H