Cylinders Flipping on their Axis

[sumayo]

We have an Assy Model with Four Cylinders. Barrels and rods are placed into the Assy separately and Constrained.

We want to do Position Reps in 3 different positions. So the driving Cylinders 1&2 are Constrained together. The other two (3&4) along with their appropriate Clevis constraints only have their barrel & Rod axis constrained ie the Cylinder Rod Axis to the Cylinder Barrel Axis of each.

Positions one and two work fine, the trouble is when the Cylinders are put into the third position, Cylinders 3&4 flip on their Axis ie they point away from each other.

has anyone had a similar problem and is there an easy fix?

[Josh_Petitt]

can you post a screen shot of positions 1, 2 and 3 in the correct positions?

When you switch to position 3, are to switching from a position that is "close" to position 3?

[Anonymous]

I have had the exact same problem, and no I don't think there is an easy
fix. I had to play around with the assy for a long time until I had the
right combination of constraints that kept it from flipping. That was on
R2008, I have not tried on R2009

"sumayo" wrote in message news:5936208@discussion.autodesk.com...
> We have an Assy Model with Four Cylinders. Barrels and rods are placed
> into the Assy separately and Constrained.
>
> We want to do Position Reps in 3 different positions. So the driving
> Cylinders 1&2 are Constrained together. The other two (3&4) along with
> their appropriate Clevis constraints only have their barrel & Rod axis
> constrained ie the Cylinder Rod Axis to the Cylinder Barrel Axis of each.
>
> Positions one and two work fine, the trouble is when the Cylinders are put
> into the third position, Cylinders 3&4 flip on their Axis ie they point
> away from each other.
>
> has anyone had a similar problem and is there an easy fix?
>

[Josh_Petitt]

>no I don't think there is an easy fix.

yes, AFAIK there really isn't an easy fix. A way to do this is to create intermediate positional reps and make "small" steps in the motion. For instance if a part revolves from 0 to 180 deg and drives a linkage, then make positional reps at 0, 45, 90, 135, 180, and don't go directly from 0 to 180. go from 0 to 45 to 90 etc.


or like Troy said, figure out another set of assembly contraints that gives the right motion.

[sumayo]

Thanks to all for replying.

Inventor seems a bit flakey here.