I created an assembly of parameterized parts (one of the parameters is actually a scale factor).
I see that when I change the parameter corresponding to a scale factor, some dimensions are scaled but their direction is reversed. Could someone explain this to me please
2°) Are all the parts drawn under the same version of solidWorks or do you have imported parts (such as screws or part imported from a supplier in the surface). It is not uncommon to have surprises with supplier parts because often the shape recognition system uses curious reference systems and often with very few or no dimensions at all. So scaling doesn't like that.
Otherwise post your ASM (using "take-away composition" that we look at the rooms that put the souk. Or post one of the pieces that puts the brain.
For my part, I always use a point of origin of the dimensions that is far from the volume (e.g. -500;-500), or in the center of the room. This usually avoids this kind of problem. On the other hand, you have to do "calculations" to subtract this dimension (500mm) in each of the dimensions... Sometimes the dimensions are reversed just because it's more "logical" for the SolidWorks algorithm.
"Sometimes the dimensions are reversed just because it's more "logical" for the SolidWorks algorithm
In fact, it seems that this is the reason for these inversions because if I change my scale setting little, everything goes well. On the other hand, if I vary it a lot (factor 5 or 10 for example), that's where the problem appears.
Do you know if it is possible to give meaning to a coast (a kind of vector depends on x, y and z if you will)?
No, I can't do a simple scaling because I have an assembly of sheet metal parts whose parameters are independent of this "scale factor". It would be too easy otherwise. :-)
However, these rib reversal problems are really disabling
The answer from @irenee.dubail makes me think that it might be enough to change the location of the origin, so that, whatever the scale factor, the sketch always remains on the same side. Too bad if the room is a little symmetrical, you just have to recreate a median plane if necessary.
Unfortunately, putting a piloted hill on the other side doesn't change anything. And putting one on each side creates overstress.
@ stefbeno
On some simple pieces, I understand that it can work. I apply this principle to my pieces to a certain extent, but I can't do this for all my ribs. That said, by changing the way I apply my ribs, I've managed to limit this problem. Generally speaking, if I don't vary my "scale factor" parameter by more than 15-20%, I don't have a rib inversion.
In short, if Solidworks could be enriched with a function that allows you to impose the direction and direction of a hill, it will be really great.