These are not asked frequently but “FAQ” seems like a good page title!
Q. How is the Windows XP version different from the normal version?
A. The XP version does not use 3D graphics acceleration to improve the quality of the graphics on screen. There are no other differences except for the “Background” feature that is unavailable for XP users. The XP version is not old software and does not need to run in compatibility mode.
Q. Are there ways to measure forces, accelerations, and other physics stuff in the software?
A. No although it is possible to export the positions and speed of selected connectors during the simulation. The exported CSV data can be used to calculate acceleration. Forces are not available because there is no way to set the mass, center of mass, etc, for any links in this software.
Q. Springs?
A. No. There is no simulation of floppy mechanisms that rely on physics of any sort to run. There is no gravity. There are no springs. There is no way to have a sliding connection slip until a limit is reached before something is moved in the mechanism.
Q. I need help. I have a problem. It’s broken. What do I do?
A. Send me an email. My email address can be found in the software in the “About” box (click the little (i) icon in the upper right corner of the window).
Q. Can I get the source code?
A. Yes if you ask me about it and have a reason for getting the source code other than to sell my hard work for your own profit.
Q. Where is the Ratio button?
A. It was moved to the Align menu. The text below the dimensions text box just shows what type of information is in the box.
Q. Where is the Connect button? (And why does the “L” key not create sliding connections?)
A. The “Connect” feature has been changed to be the “Link” feature. Use the “Link” button to create a link from selected connectors. The “C” button still works because there’s no other use for it. The “L” button now creates links this way and is no longer the keyboard shortcut for creating sliding connections.
The following copy paste into notepad and saved as a linkage file, does not give the result I want, in design mode, the slider follows the cam, in running mode, not.
It should work as a four bar mechanism, where the cam serves as a corner, thus pitching the outer link, in or out, since one side of the outer link is fixed to the anchor, and the other side, to the actuator which serves as a telescopic bar, following the cam. (final purpose is simulating the pitching of a vertical axis wind turbine)
Hello Dave
I am having trouble installing Linkage. I am running Windows 11.
The installer MSI works OK and places software in “program files” and a shortcut on the desktop.
When I double click on the shortcut, it gives me a “wait” icon for about 5 seconds, and then nothing more happens.
I check in the task manager, and Linkage is now loaded, but it is not coming to the screen.
I also tried running Linkage directly from it’s install directory using “run as administrator” and the same thing (nothing) happens.
I do a complete removal of Linkage after all these attempts.
Anyway suggestions are welcome.
Thanks, Geoffrey
just reinstall, works fine for me
Hello from New Zealand,
Just want to say a huge thank you for this software. I do a range of mechanical modelling in various CAD programs, but have to say linkage it always the first thing I open to sketch an idea. It is so so powerful to quickly prototype in 2d.
I often copy and paste a design, make a modificaiton and then repeat to visualise a range of options.
I notice if I use ‘Rotating Anchor’ with an oscillation start angle – then the start angles tend to reset to 0 if I make modifications to the structure. Is there anything I can do to stop this happening.
Cameron
Hi Dave,
I just discovered Linkage (from a post in the Rhino3d forum) and have leapt in with an attempt to simulate the linkages of a 155 degree kitchen cabinet hinge. There are five links in a single chain between two anchors. I get that having a chain of more than two links between anchors generally introduces indeterminacy, but in the case of these hinges we know by observation that there is only one viable transformation. Is there any way that Linkage can be made to simulate this?
There is a difficulty in simulating links when the link is positioned based on more than two other links. I have yet to find a mathematician who can tell me how to solve the problem although one suggested nine simultaneous equations. You can read a bit about the problem here: https://blog.rectorsquid.com/assur-triad-simulation/
I suspect that you are seeing a similar issue as with the triad. I am encountering these problems more these days and it sucks that I can’t solve this for more than the simplest triad situation.
So it’s not really indeterminism that is the problem, it’s a flaw in the underlying simulation algorithm that can’t handle three or more variables in the positioning of links.
I have a few posts here and there about how the simulator works. It just uses circle-to-circle intersections to find link positions. For sliding connections, there is a special piece of code for each variation but the code still uses computations that could be done with a straightedge, compass, and maybe a ruler in some situations.
I am teaching Linkage program to teach a Kinematics course this semester. I am explaining the basic mechanisms using Linkage and let students do the mechanism simulation using Linkage. It helps students understand the theories and intriges students about the course. Thanks.
If only Linkage provides the functionality to simulate cam mechanism (it doesn’t have to be complicated cam, maybe just circular cam with offset center), it will be perfect for the course.
I am working on cams that will be well suited for mechanical toys and automata but work is slow going and there is a lot of work to do to the interface to allow editing of Bezier curves. So far, All I have is the ability to drop in a closed polyline (with Bezier curve segments and straight segments) and select it to rotate and stretch/scale it. Node editing is complicated and will have snapping capabilities to align with other elements and other features like that. I have the simulation code worked out, which is very tricky with Bezier curves, but seems like it will be quick and accurate to the level needed for this software. Unfortunately, I can only work on this a few hours a week right now due to other job pressures and I can’t estimate when it will be done. It won’t be ready this semester, I’m sure of that.
if there’s anything else I can help with, please don’t hesitate to let me know. My email address is in the documentation and in the “about” box in the program.
Have you, or do you know anyone who has, created a version for Mac (not using parallels)? Or perhaps a web version?
At this time, no one has been able to port the code. It’s tens of thousands of lines of source code and once anyone sees how much there is, they give up.
I started to write a Mac version from scratch because it would be written so much better the second time around. I could not find a decent way to render a few hundred lines and circles in a window without my 5 year old Mac slowing down. I dread having to use the lowest level of 3D graphics accelerator for the rendering. I also have some new features I want to write more than I want to write a Mac version (like cams and curves that I’m working on now).
Just updated to v3.11.3 – Went fine.
Tested out the DXF export on the old ‘straight-line-lift’ model.
Into ACAD and the geometry and size came in fine. Only around .0003″ difference – rounding error I’d call it.
Should do fine to figure out the gearing.
This is still the best available software on the internet! I have been working on a running horse which you can see here. https://www.youtube.com/watch?v=6Qo5z8EnsRE
At the moment I am adding extra connector points to basically try and draw the shape of the piece. This helps me to visualize what it will look like. Would it be possible in the future to be able to have the connector points connect to each other to create very rudimentary shapes. Almost like drawing dot-to-dot whilst having the mechanism still work the same way. I am trying to teach myself Solidworks but it just takes so much time to model. If your software could draw really basic shapes I would be willing to pay!
And the updated running horse in action! https://www.youtube.com/watch?v=wI_gKDzs26w
That’s awesome.
Hi Dave,
I found your site by doing a search for hinge design software to help me aid in designing a hinge for a hood of a car. It needs to swing up about 210degrees and produce a lift of 8 inches. The real problem is it has very shallow constrains and I’m wondering how I can specify the constrains and surfaces of the stationary and moving part with limits so it would clear perpendicular surface.
Thanks Jerry
Jerry, the Linkage program is a rapid prototyping tool and there is no way to specify constraints like you describe. You would need to make tiny visual changes to the mechanism until it works the way you need. In the few cases where I needed very precise end points to the movement of a link connected to a lever, I zoomed in on the end of the lever and then used the arrow keys to move it a very small amount at time. You can also select two connectors and enter a percentage in the text box in the tool bar and when you press ENTER, the second connector will move so that the distance between the two is adjusted to that percentage. Small values like 100.001% are possible and might allow for manual adjustment of the mechanism to get it to work.
Feel free to send me an email with a description of what you think the Linkage program should provide and how it would work. I can at least keep a record of the request and maybe add something eventually.
Linkage will Replace Force Effect Motion; Autodesk left people hanging.
Hi.
I am very fond of Linkage software. Thank you Dave.
I am interested in robotics, especially in quadruped. If you allow me, I will suggest some ideas.
You can see few linkages videos in my YouTube channel, https://www.youtube.com/user/oracid1/videos
Your work is outstanding. I noticed a comment on one of the videos about the mechanism being sensitive and you are very right. If any link is not exactly the right shape, the mechanism could bind and not function. In the past when I have connected to straight line approximation mechanisms together, I had to stick in a very tiny connector link to account for the imperfect line created by the types of mechanisms. Getting an interesting mechanism to not bind can be tricky.
I subscribed to your channel so I can keep up on your work.
Hi Dave.
Thank you for your compliments, but it’s thanks to Linkage that I can do this kind of thing. So, thanks to you.
I have just posted a video, https://youtu.be/h0Xz_barINw , so I can explain some idea to you, if you allow me.
To understand my goal, you can see a lot of videos in this playlist, https://www.youtube.com/playlist?list=PLtVavDLiwv7F38lxOfDJEup01BYlxhOJa .
A good example is the 24th video.
As you can see in my video, there is 2 servos. One on the left and one on the right.
1 – Is it possible to make an angle tool which follow the bars to which it is joined ? I know there is already an angle tool, but it does not follow the bars.
2 – Is it possible to retrieve the different values of this angle in a file ? And why not, to see the value in a little window.
3 – Is it possible to give a liste of (x,y) couples so that the end of the linkage will follow ? Example : lets get the (x,y) couples of the Theo Jansen trace, then ask to your end linkage to follow this list of couples.
Of course, I know this is a lot of work, but I am sure that if you will do this kind of things, Linkage will become a standard in robotics.
I am French and my English is not good. I hope you understand all what I mean.
I hope you don’t mind I write to you in this page. If you better like, we can use email.
1. If you drop in an angle measurement, you can then “fasten” the three points to three connectors in your mechanism. The angle shown is dynamic and will change as the points of the angle measurement move.
2, 3. There’s no way to get the angles in a file or a way to dynamically create a mechanism or have it follow a path.
You can enable “drawing” of a connector and then once the mechanism goes through its movement, you can export the drawing coordinates. You cold do that and use a spreadsheet to compute the various angles. That might help with any angle work you are doing.
Can you tell me more about why you need the angles in a written form? I can guess that you need to see the minimums and maximums because it affects how the parts are built. If there is some other reason, it might help me figure out what the software could do to help.
Dave
In Win 7 (64) it installs and loads fine (seemingly), but the parts only appear on the screen fleetingly and when you click around. Even the sample mechanisms don’t show up correctly. Any thoughts?
Cheers
Some computers have trouble with the graphics. The Linkage software uses the 3D accelerator card for the smooth 2D rendering. If the 32 bit version causes the same problems, try the Windows Xp version. The only difference in that version is the graphics drivers and all that is missing is the smoothness of the lines.
1- I am so sorry. I have to progress in using your Linkage software. It is even far better that I could imagine.
I have selected the “fasten” button and the angle follow very well the bars.
2- This is very great that I can export the coordinates of each points. Yes of course, I can calculate the angle if I have the coordinates of a point with a 2 bars linkage. But it will be far better to have the both, angles and coordinates because it is not so easy with a 5 bars linkage to calculate the angles of the 2 servos positions, as in my last video.
I don’t know how to explain.
4 – Let’s say I want to draw a ligne. So I can connect the end of my 5 bars linkage to an horizontal linear actuator with 2 anchors so to draw a ligne. I did not succeed. It does not work.
5 – Let’s say I want to draw a circle. The same, I must connect the end of my 5 bars linkage to the circle with a slider. I don’ even know how to do.
The goal is to connect the end of the 5 bars linkage to a path, like a Theo Jansen path. So that the end of the linkage must follow this path.
Sorry if I am not clear.
Hi, Dave!
The programe is very cool.
But, I can not understand how to do gear to gear rack.
I would like to say that the XP-version works in Ubuntu via a Wine without problems.
It can be useful.
Dave, that do you think about FreeCAD, and Linkage like addon?
The Linkage program does not support rack gears. It only handles circular gearing.
That’s great that it works with Ubuntu. I tried using Wine and a Macintosh and didn’t have great results. Using Parallels desktop for Mac did work, even with the non-XP version, which was just a little surprising given that the non-XP version uses the 3D accelerator for it’s smooth graphics.
A Linkage add-on for FreeCAD would be a whole new project and probably more than I can take on now or in the future.
Hi Dave,
This is one of the coolest programs I have seen in a long time. I have been building simple linkages and making custom prototype parts from 1/8 and 1/4 thick acrylic for some time and this program will allow me to make much more complex mechanisms. I think this will be a good tool for our High School Robotics club as well.
Thanks, Craig
Craig, I’m glad the software is useful to you. Feel free to let me know of any comments, suggestions, complaints, etc., you might have. You can send me email using my email address in the Linkage “About” box if you want.
Good Day Dave
Awesome program. I am designing a link, but need to advance a gear,
on the similar way you would advance a “rachet”.
I am using a sprag clutch, a one-way freewheel clutch in my project,
is it possible that your program can simulate such a design?
Regards
Franklin
Sorry, the software can only simulate pivot and sliding connections that have positive connections to the other parts. A ratchet typically uses a spring and has an intermittent connection to the wheel it pushes or pulls.
Hi David,
I’m having difficulties installing linkage. I keep on getting an error message saying that another installation is in progress while actually I’m not installing anything while installing linkage. What’s going on? Anyway around this issue?
None of the other Linkage users have reported this problem. The Linkage installation uses Windows to do the install and I have no way that I know of to cause this sort of problem. I would suggest restarting Windows as the first thing to do, if that’s possible for you. You can also open the Task Manager and look for a process called msiexec.exe. I usually seen either none of these or one of these. During an install of the Linkage software, I see a second msiexec.exe with a command line that references the Linkage installer file. Maybe Windows is installing something from MS and it’s stuck.
You can download the Linkage executable file and documentation and run it without using the installer. Just create a folder or directory somewhere in the file system (on the hard drive) and copy the files there. Then double-click linkage.exe to run it. The executable and documentation can be found here: http://www.rectorsquid.com/linkage.zip
Dave
Hi David,
This is awesome software.
However, the current Lock function locks the link lengths but also the capability of placing the anchors to exact grid locations. Unless there is a special sequence to follow.
The software also locks up when one tries to move a common joint belonging to two locked links.
As it is, it is great to quickly design a mechanism and see it working, but it would be great if you could fix this locking problem.
It would be also great if the dimensions of locked links could be overwritten from the Dimension box. And sometime locked links’ dimensions can be dragged when they shouldn’t be.
Cheers,
Murat
I’m glad you like the software. I am working on a lot of bug fixes and improvements right now. I’ll work on the locking and see if I can fix the lock-up bug and the other problems you mentioned. I’m not sure about letting the locked links be changed by typing dimensions – they are locked and maybe that’s a violation of the lock. I’ll think about it. The locked links should certainly not change when dragged in any situation though.
Locked links can be dragged so that should allow the anchors to be snapped to the grid. I’ll test that too, and make sure it works if there is a bug.
Thanks for the feedback. Feel free to find my email address in the “About” box or on this blog and send me any more feedback using email. Or use the blog comments, whichever is best for you.
Dave
Murat,
I just created two links and joined them at a connector. Then I selected one and clicked the “Lock” button then selected the other and clicked the “Lock” button. When I drag the connectors or the links, it all works fine. Then I changed one of the connectors to be an anchor and it all worked fine. Finally, I locked the anchor and then the problems started.
I think that locking an anchor does what it should and keeps it from being moved. If you want to move it, unlock it.
Keep in mind that if you select a link by drawing a selection box around it instead of just clicking on it, the connectors are also selected. Locking the link using the “Lock” button should then also lock the connectors and cause the sort of problem you see. That also causes the hang when you try to drag the locked links. I’m working on fixing that bug right now.
Dave
Hi Dave,
Sorry for the late reply. Thanks for trying to fix the issue. I decided to give another go after few years :). The locking inconsistencies are still lingering, such a locked anchor moving to accommodate a joining even while one of the links isn’t locked and can be stretched instead. I think that ideally there should be no stretching of any linkage when joined to another one, the result should be to move the free link to meet the joint. In more general terms, the joining should solve for the positioning of all the related positions on the fly, provided there are enough DOFs left unconstrained. No locked anchors should ever be moved to accommodate any action. For a more intuitive accurate assembly, may I suggest the following:
1. Once locked, and anchor should not move under any circumstance.
2. Locking of a linkage should only lock its geometry, and depending on the existing BCs already imposed upon it, it should be free to move about, or rotate about any prevailing joints without stretching. This implies that the solver should work on the fly to solve for its position. This feature could be so that it solves for the positions of all affected links as the selected link (or one of its joints) is moved around.
3. Any link should be able to be positioned using polar coordinates, as well as in cartesian coordinates. The cartesian coordinates should allow to enter one coordinate, and calculate the other for a locked link length. An individual lock next to these entries would allow implicitly calculating the other parameters.
Happy to give more suggestions. I think that this is a great tool that I would be happy to pay for once the above are resolved, as I would like to use it in my Mechanics of Machines class as a teaching tool without frustrating my students :).