First off, if you’ve never heard of a Wanzer Needle Bar, you can see an explanation here: Wanzer Needle Bar
I made this animation of a Wanzer needle bar using Pro/Engineer for an assignment this fall - http://www.ualberta.ca/~twgreene/WANZER.mpg. Today I tried to animate it with Blender, but failed utterly. I would like to set it up so that the corners of the the green linkage move along the slots automatically when the wheel is turned. Can someone more familiar with Blender give me some suggestions or tell me if this is possible? Here’s my .blend if anyone wants to look at it - http://www.ualberta.ca/~twgreene/wanzer.blend. The mesh is a mess because I didn’t clean it up after importing from Pro/E.
Update: I gave up trying to set this up with constraints. I briefly played with using three hooks to move the corners of the green bar around, but I couldn’t get the hooks to move right, and wasn’t sure how to make them move the bar without deforming it. Eventually, I just put a keyframe for every 15° the wheel turned and moved the bar manually. It turned out okay. Here’s the .blend: http://www.ualberta.ca/~twgreene/wanzer_keyf.blend
Can anybody think of another way to set this up? My keyframe method isn’t very elegant.
I’m also wondering if I’d get a better response on the general board. This one doesn’t seem very active.
I’ve been looking at this mechanism and its movement and think its looking realy interesting. I have tried to similate the movement with constraints but have not succeded (yet). It looks like some kind of cardioid (see link). Do you have a more mathematical explanation of this Wanzer needlebar movement somewhere?
Hi, Thanks for the responses; I’m glad this topic is of some interest.
SoftWork - I believe both boints of the triangular bar that are on the wheel travel in cardioids. I’ll be at school tomorrow, so I can try to get the trace curves of those two points from Pro/E if that will help you. Unfortunately I don’t have a mathematical relationship for the movement, and all a quick google search turns up is that it was invented in the late 1800’s by a Canadian sewing machine maker named Wanzer. I’m in the middle of finals right now, so I don’t have time to work anything out. I may be able to get to it sometime next week, but I can’t promise anything over Christmas. I hope you can get the constraints working. I’ve never used constraints before so I didn’t know what I was doing when I tried.
S68 - Wow, I’m honored that you’re giving this project a try. I’m a big fan of your gears script. Good luck.
I’m also giving it a try. I already know how to animate this cardioid with two circles. Only after I keyframed the position of the circles and press Alt-A to animate, I get some wierd movements. That’s taking the most time to figure out right now. I’ll let you know when I have solved this (if S68 didn’t solve it before me…).
Thanks for giving it a shot. Using two circles to make the cardioid sounds like a great idea, I hope it works out. I’ll be very interested to see your results if you get it to work.
Nice Teeth
Looking at your simple solution I realise I have to improve my animation skills. I had modelled about the same only didn’t come up with the cyclic extrapolation. Good job.
Well, as promised, I have the trace curves for the three points of the triangular piece.
The two light blue cardioids are the path those points take as the wheel completes one revolution. The third point stays on the horizontal dark blue line.
If you look closely, you can see that the cardioids are neither symmetrical nor centered on the wheel, which unfortunately makes this problem more complex. It also might account for some of the erratic movement you were seeing SoftWork.
I also have a spreadsheet with the X and Y positions of each of the points plotted over time. This is the closest I can get to a mathematical explanation of the movement right now. You can download the spreadsheet in Excell format here -> http://www.ualberta.ca/~twgreene/needle_bar_motion.xls. All the positions have been translated so that the origin is at the center of the wheel. I tried to fit some cosine curves to one of the Y positions, which you can see in the sheet, but the motion was a little too complex.
Don’t feel bad if you have to give up on this SoftWork; I think Blender needs rigid body kinematics to do this kind of animation. I made the animation in the first post in just a few minutes using kinematics in Pro/E.
I’m sorry you couldn’t get it to work S68, maybe the new animation tools in 2.37 will include something that will make this easier - or possible.
They would be cardioid if the sliding straight path were passing through the circle center.
One of the two looks, to the eye, a cardioid because that sliding path is quite close to the center, but the other curve is clearly asymmetric, which reveals it is not a cardioid, mathematically.
I haven’t got time to try the script, but I’ll try sooner or later. Looks an interesting problem!
I think you could come up with a pretty close approximation of the movement (of the device in TrevorG’s latest post) by animating the circular rotation of the wheel as one step, and the horizontal movement separately.
But another, quite reasonable approach would be to calculate all of this stuff ahead of time, write out the vertex positions and the rotations to a constant table in Python format, then write a short Python function that simply looks-up and outright assigns the correct position and/or rotation of each of the necessary objects. No (further) calculations required.
