Rod component

Dear Daniel (et al.)

Just for fun, I tried modelling this beautiful clock design by Matthew Gilbert in K2:

Unfortunately, when the structure is peturbed in my own model I can't seem to maintain the bending resistance of the short three-point polylines using the rod component in K2:

They seem to resist no bending at all, although the structure (as constrained in the video and in the attached definition) should have only 1 DOF.

Please find attached definition, and any help and advice is appreciated. I seem to get this running in K1 with the 'rod' userobject equivalent, but not in K2. It's probably user error of course, but would be nice to find out!

Thanks in advance,

John.

Attachments:

KangarooTeaser_Clock_K2.gh, 15 KB

Replies to This Discussion

Permalink Reply by jjsolly on September 17, 2017 at 5:06am

Hey John,

Two things jump out:

Relationship of scale - the elements in your model are quite large compared to the default strengths. Daniel would be able to provide a better explanation of why that affects things than I can. But I got better results by bumping up axial and bending strength and scaling the whole model to 0.1 of itself.

More importantly - I think your model is actually overconstrained so it can't move as you want. The pairs of members that form a single rigid part need to have the same length to work and your ones don't, the outer ones are in fact longer, which locks the whole system in place. So I fixed that and got the below:

Also it was an opportunity to play with Dan's new rigid point groups goal. Which seems to allow even more stability and faster movement in this situation.

Attachments: