| Author |
|
enginerd
Member
Joined: 21 February 2006
Online Status: Offline
Posts: 3
|
| Posted: 08 August 2009 at 12:42am | IP Logged
|
|
|
I am trying to design a 4 link mechanism with a coupler
point. There are defined constraints for the coupler
curve, it has a lateral constraint and has to raise to a
defined height. There is also a constraint on the initial
height and overall length of the mechanism (starting
location of the coupler point). I have created loop
closure equations relating all 4 links and angles. This
seems next to impossible to figure out by hand or try to
write a program to do it. How can I get BMX to figure out
the proper link lengths to fit the given parameters? Is
it possible to do this?
Just so you know, I have already made a best guess
skeleton mechanism and created several measurements
relating the coupler point to its start point, angle
travel, etc. Motion analysis features for the lateral
movement, vertical and angle measurements have also been
created.
Please let me know if a drawing or assembly file is
required for help.
Thank you
|
| Back to Top |
|
|
Mloew
Moderator Group
Charter Member
Joined: 10 December 2002
Location: United States
Online Status: Offline
Posts: 163
|
| Posted: 08 August 2009 at 11:21am | IP Logged
|
|
|
This is a GREAT problem. Dr. Andreas Vlahinos and I developed a method for optimizing mechanisms using BMX and MDX. This work was presented at the 2004 PTC/USER World Event.
Basically, you want to create an objective function that can be minimized. In this case it is an error function abs(current distance to desired location). Let BMX minimize this error function and by altering the lengths of linkages (or locations of hard-points) subject to your constraints. Please let me know if this makes sense to you.
Be careful, however, optimal solutions can be fragile. Small tolerances in the locations or lengths may destroy the eventual desired solution. This is an entirely different class of problems. This is a current area of interest for me...
__________________
Best regards,
Matthew Ian Loew
www.linkedin.com/in/matthewloew
|
| Back to Top |
|
| |
enginerd
Member
Joined: 21 February 2006
Online Status: Offline
Posts: 3
|
| Posted: 08 August 2009 at 12:18pm | IP Logged
|
|
|
I already have the function for the location of the coupler
point. It is based on the crank, rocker and coupler
vectors. I wouldn't know how to enter it in to ProE though
since it is in complex algebra form. Can ProE handle
imaginary numbers? Or do I have to translate it into an i+j
vector?
|
| Back to Top |
|
| |
Mloew
Moderator Group
Charter Member
Joined: 10 December 2002
Location: United States
Online Status: Offline
Posts: 163
|
| Posted: 08 August 2009 at 7:31pm | IP Logged
|
|
|
Pro/E can't handle imaginary numbers explicitly, but it can handle quite easily a linear system of equations.
I had started a FAQ on the eng-Tips.com site:
Mathematical Operators used in Pro/E Relations
__________________
Best regards,
Matthew Ian Loew
www.linkedin.com/in/matthewloew
|
| Back to Top |
|
| |
enginerd
Member
Joined: 21 February 2006
Online Status: Offline
Posts: 3
|
| Posted: 08 August 2009 at 9:45pm | IP Logged
|
|
|
Thanks for the help, i'll see what i can do with your
suggestion and what I have already.
|
| Back to Top |
|
| |
Active Topics
Search
Help
Register
Login
Topic: Four Link Mechanism Design Using BMX
