hinge ik chain, any advice welcome

[Stroker]

Something is not right.

You have 3 different fixed points that define a triangle (you even have a 4th one tossed in there). If all 3 of those points are fixed with no sliding, then nothing will move.

If you take out the short piece, it should work just fine.

Or am I missing something?

[santiago]

Hi Stroker,
if there were just three joint forming a triangle then it sholdn't move, but since there are four, then it can behave like a square that can be turned into a rombus, and made completely flat. Think of a box with no top or bottom, it can be folded to become flat.

When I mean fixed joints, I mean that they don't slide on the arm they are attached to.

[santiago]

Here's another image to show what this hinge looks like when it's half-way open.

http://www.santiagodiaz.com/gallery/xtra/hinge-joints2.jpg

This is a poor sketch but it gives the general idea.
Joint A never moves from its place.
The distance AB + AD is equal to the distance BC + CD

As the angle between the Mainframe and the vent frame becomes smaller, the joint E slides downwards in this point of view.
The little black arrows show the direction these joints move as the hinge closes.

I hope this helps at all

[Stroker]

Thank you for clarifying.
I'll take a whack at it later, unless someone beats me to it.

edit:
I did some poking around and this looks like a perfect candidate for a Reactor Controller. Unfortunately, I'm not proficient with Reactor Controller. I've been meaning to learn how to use these.

[Stroker]

I did things kind of weird, but I think I'm getting the hang of it.

Only did the one set of hinges to the manipulator. Doing the other hinge shouldn't be hard with basic IK and and some creative linking.

windowhinge.zip (http://www.playfiddlelearn.com/stuff/windowhinge.zip)
- Max 5.1 ~12k

Take a peak at the reactor controller from the motion panel for the various bits-n-pieces.


Or how about reactor pin-nail or nail-points or whatever those things are called? I think this would be the better solution. But, again, I'm not proficient with these things. Hmmm...

[Stroker]

Here is one that uses 2 path contraints and an IK chain. Wired to a slider. I only did the one bar because adding the rest should be a breeze once the long arm is moving properly.

Now, it's a real slop job, but it looks awlfully good for how quick I slapped it together. I'm sure you can get as precise as you care to. You know, math and stuff.

windowhinge2.zip (http://www.playfiddlelearn.com/stuff/windowhinge2.zip) ~13k

It's been fun and a mild learning experience.
Thanks for that, Santiago.

edit:
LOL! I just realized that the second one is a little bit wrong. Still, the idea is sound.

[santiago]

Hey Stroker, the path constraint sounds like a great idea!
I never thought of that one, I'll apply it to the sliding joint first, I think I might not need to add the curve path constraint.

[Stroker]

Oh, man. This is the most fun I've had in awhile. Still a tad of wiggle in there, but it is very close.

windowhinge3.zip (http://www.playfiddlelearn.com/stuff/windowhinge3.zip) ~15k

Have fun tearing it apart, because I don't think I can explain half the stuff going on in there.

Man, that was fun.

If you come up with a more elegant solution, I'm all ears.

[santiago]

I think I'm on to something Stroker, hopefully tonight it'll be ready...

[santiago]

By the way, this is the first time I've seen a "swivel angle", it might take me some time understanding how it works.
Your rig is quite complicated
I rotated the base object and discovered that the end effector on the vent frame arm doesn't follow.

I've managed to make the hinge work in a slightly different way than yours, I haven't attached a manipulator yet, but I'd like to.
Once I get something fixed. Something that helped me alot from your rig is the path constraint and linking sequence which I found perfect for this kind of setup.

I can't seem to upload to my server space, can I email my files to you?

They still have an issue, but I think it's minor and you could probably help me with it effortlessly.

[santiago]

Ok, I uploaded the files:

version 1 of hinge (http://www.santiagodiaz.com/gallery/xtra/hinge - path constraint 2 - working.zip)

version 2 of hinge (http://www.santiagodiaz.com/gallery/xtra/hinge - path constraint 3 - working.zip)

[LFShade]

This is exactly the sort of setup that the reactor controller excels at. I've used your geometry to set up a rig using this method. You can fold the hinge by sliding the red point object, or by using the slider. Note the reactor controllers on the Z rotation channels of ARM-AB and ARM-BC. They are triggered by the percent-along-path of the red point, with a number of triggers due to the nonlinear relationship between the sliding and joint rotation. The hinge point corresponding to D in your above drawing still doesn't track properly, could this be the result of a geometric inaccuracy?

hinge_reactor.zip (http://www.scriptspot.com/rhyde/files/hinge_reactor.zip)


RH

[santiago]

LFShade
It looks almost perfect. Possibly the geometry failed somewhere during max modeling. In Autocad the lengths of AB + AD are equal to BC + CD.

I have another hinge that is supposed to work similarly, but I haven't done anything to it in max yet. I'm posting it so that you can verify that the dimensions do match.

hinge reactor with friction hinge (http://www.santiagodiaz.com/gallery/xtra/hinge_reactor with friction hinge.zip)


I'll try to fix the Joint D issue, although I get the feeling that perhaps a dimension could be off by a couple thousandths of an inch, which might cause that shifting around of that particular joint instead of staying glued onto the CDE arm.
But the friction hinge should work fine.

[Stroker]

Good to know that my first instinct to use a reactor controller was seconded. Still, I had a blast making that last convoluted rig. Talk about a mess.

I'm walking away a little smarter. Thanks.
I'm going back under my rock now.

[santiago]

I'm trying to recreate LFShade's reactor setup but so far am unsuccessful.
Here are two files, the first is the setup with the vent frame and main frame in place. The second s my attempt to recreate the reactor setup:

http://www.santiagodiaz.com/gallery/xtra/hinge_reactor w frames 2.zip
http://www.santiagodiaz.com/gallery/xtra/hinge_reactor w frames 2 working separately.zip


My main question is how the two arms with reactor controllers (AB and BC) are reacting to a "Percent" value of eachother, when they don't have this value, or do they? I can't find it.

[LFShade]

They are not reacting to each other, but to the percent value of the red point's path constraint controller.


RH

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.