Dynamic simulation model moves while setting up joints

[pachubby]

Hi everyone,

Hopefully you guys can answer my novice question and explain what's happening in the dynamic simulation environment.

As I am inserting the last joint on a 4-bar linkage, the assembly re-orientates by X degrees through the final degree of freedom. This is frustrating because I entered the DS environment with the assembly model positioned at one extremity of movement, and now I have lost the starting position. I can't quite work out why the software has picked this number of X degrees.

At a wild guess I might assume one of the earlier joint co-ordinate systems is defining a global position of everything else.

To explain my models: This 4-bar link looks like a simple diamond, two long side, two shorter sides. The base link is grounded, the next two links can each rotate using revolution joints, the 4th link can then translate and is connected to the ends of the 2nd and 3rd links. Mechanism status shows 1 DoF. There's also a pair of compression spring style mounts, attached to opposing corners of the diamond shape. In total there are 4 revolution joints, 2 point-line joints and 1 prismatic joints. Essentially I want to control the position of the 4-bar link w.r.t. a perpendicular distance from a flat surface on another grounded part.

If you would like the model uploaded, I’ll have to create a non-IP version, but that shouldn’t take to long.

I know that I can edit a joint’s properties and update the initial conditions to choose a starting position I want. But I would like to understand the reason why it moves by X degrees.

FYI - I have been teaching myself DS for the last few weeks and have been using Wasim Younis’s book (massive thumbs up and kudos to Wasim) and completing the different tutorials. So my general understanding of DS is much improved, but as to the reason why the mechanism moves by X degrees has left me totally scratching my head.

For reference I am using Autodesk Inventor 2019, build 330, release 2019.4.

Looking forward to some insight!

[pachubby]

Hi everyone,

Hopefully you guys can answer my novice question and explain what's happening in the dynamic simulation environment.

As I am inserting the last joint on a 4-bar linkage, the assembly re-orientates by X degrees through the final degree of freedom. This is frustrating because I entered the DS environment with the assembly model positioned at one extremity of movement, and now I have lost the starting position. I can't quite work out why the software has picked this number of X degrees.

At a wild guess I might assume one of the earlier joint co-ordinate systems is defining a global position of everything else.

To explain my models: This 4-bar link looks like a simple diamond, two long side, two shorter sides. The base link is grounded, the next two links can each rotate using revolution joints, the 4th link can then translate and is connected to the ends of the 2nd and 3rd links. Mechanism status shows 1 DoF. There's also a pair of compression spring style mounts, attached to opposing corners of the diamond shape. In total there are 4 revolution joints, 2 point-line joints and 1 prismatic joints. Essentially I want to control the position of the 4-bar link w.r.t. a perpendicular distance from a flat surface on another grounded part.

If you would like the model uploaded, I’ll have to create a non-IP version, but that shouldn’t take to long.

I know that I can edit a joint’s properties and update the initial conditions to choose a starting position I want. But I would like to understand the reason why it moves by X degrees.

FYI - I have been teaching myself DS for the last few weeks and have been using Wasim Younis’s book (massive thumbs up and kudos to Wasim) and completing the different tutorials. So my general understanding of DS is much improved, but as to the reason why the mechanism moves by X degrees has left me totally scratching my head.

For reference I am using Autodesk Inventor 2019, build 330, release 2019.4.

Looking forward to some insight!

[johnsonshiue]

Hi! It could be a bug, a limitation, or a bad setting. Without your files, it is very hard to tell. If possible, please share the files here. Forum experts can help take a look.

Many thanks!

[johnsonshiue]

Hi! It could be a bug, a limitation, or a bad setting. Without your files, it is very hard to tell. If possible, please share the files here. Forum experts can help take a look.

Many thanks!

[pachubby]

Hi @johnsonshiue 

I have attached a Pack&Go where I have been able to replicate the problem.  The attached image highlights how the model has moved (from 20mm to 19.980...mm).

My workflow:

Before entering the dynamic simulation (DS) environment the assembly was positioned how I wanted it (for the purposes of this design, a neutral position, and 20mm away from a grounded datum surface). I set this with Mate4 which is then suppressed. Once I enter the DS environment I first turn off the "Automatically convert..." in simulation setting. Then manually convert constraints, I create: revolutions1,2,3&4, and point-lines1. Then I insert point-line2 and prismatic7. As soon as I create prismatic7, the assembly moves by 0.0193...mm.

I would like to know why it moves?

Do I need to change my workflow?

Or how do I control the initial position of the model? ... Note that I want to position the model X distance from the datum surface using rational (whole) numbers. Rather than changing the angular initial condition on one of the revolution joints.

Many thanks,

[pachubby]

Hi @johnsonshiue 

I have attached a Pack&Go where I have been able to replicate the problem.  The attached image highlights how the model has moved (from 20mm to 19.980...mm).

My workflow:

Before entering the dynamic simulation (DS) environment the assembly was positioned how I wanted it (for the purposes of this design, a neutral position, and 20mm away from a grounded datum surface). I set this with Mate4 which is then suppressed. Once I enter the DS environment I first turn off the "Automatically convert..." in simulation setting. Then manually convert constraints, I create: revolutions1,2,3&4, and point-lines1. Then I insert point-line2 and prismatic7. As soon as I create prismatic7, the assembly moves by 0.0193...mm.

I would like to know why it moves?

Do I need to change my workflow?

Or how do I control the initial position of the model? ... Note that I want to position the model X distance from the datum surface using rational (whole) numbers. Rather than changing the angular initial condition on one of the revolution joints.

Many thanks,

[JDMather · Accepted solution]

Examine the Attached assembly.

If you set up you assembly constraints correctly - you generally do not need to manually create Joints in Dynamic Simulation as they will be created for you automatically.

[JDMather · Accepted solution]

Examine the Attached assembly.

If you set up you assembly constraints correctly - you generally do not need to manually create Joints in Dynamic Simulation as they will be created for you automatically.

[pachubby]

Hi @JDMather

Fantastic - thanks! I now know what I was doing wrong in setting up my assembly constraints and joints.

Reviewing the assembly model you sent back in your reply, has lead me to go off and refresh my memory of 3D free body diagram mechanics knowledge. Then the eureka moment happened and the final bit of the dynamic simulation jigsaw clicked into place and I understand how to use it properly. So many thanks.

For anyone reading this post, I strongly recommend buying Wasim Younis's book, Up And Running With Autodesk Inventor Professional Part 2 Dynamic Simulation. There’s not much online resource for Dynamic Simulation on either the Autodesk website or YouTube.

[pachubby]

Hi @JDMather

Fantastic - thanks! I now know what I was doing wrong in setting up my assembly constraints and joints.

Reviewing the assembly model you sent back in your reply, has lead me to go off and refresh my memory of 3D free body diagram mechanics knowledge. Then the eureka moment happened and the final bit of the dynamic simulation jigsaw clicked into place and I understand how to use it properly. So many thanks.

For anyone reading this post, I strongly recommend buying Wasim Younis's book, Up And Running With Autodesk Inventor Professional Part 2 Dynamic Simulation. There’s not much online resource for Dynamic Simulation on either the Autodesk website or YouTube.

[JDMather]

---

@pachubby wrote:

There’s not much online resource for Dynamic Simulation on ...YouTube.

---

Hmmp.

https://www.youtube.com/playlist?list=PLp5izJt_zvN08mpY4UcYrzJV2N6QOLAIR

https://www.youtube.com/playlist?list=PLp5izJt_zvN0GnwEUFR1zE06nQvDAWxz2

[JDMather]

---

@pachubby wrote:

There’s not much online resource for Dynamic Simulation on ...YouTube.

---

Hmmp.

https://www.youtube.com/playlist?list=PLp5izJt_zvN08mpY4UcYrzJV2N6QOLAIR

https://www.youtube.com/playlist?list=PLp5izJt_zvN0GnwEUFR1zE06nQvDAWxz2

[pachubby]

A-ha! I stand corrected.

Thanks, I'll take a look.

[pachubby]

A-ha! I stand corrected.

Thanks, I'll take a look.

[AnthonyHarris90842]

Hi JD

The above playlists do not exist anymore.
Do you perhaps have a new link

[AnthonyHarris90842]

Hi JD

The above playlists do not exist anymore.
Do you perhaps have a new link

[JDMather]

@AnthonyHarris90842 

I no longer have access to Autodesk software.

[JDMather]

@AnthonyHarris90842 

I no longer have access to Autodesk software.

[JDMather]

@AnthonyHarris90842 

Split the model in half to simplify the set up of constraints (Joints) and solve for only one side.

This will give the force for one cylinder which is equivalent in the second cylinder.

Otherwise you will likely be plagued with overdefined conditions in the Joint solver.

You will have a Revolution Joint for the Rotation DoF.

Constrain the Cylinder for Revolution Joint at one end and Point-Line or Point-Point at the other end.

Constrain the Cylinder Piston such that it has a single Translational DoF.

Solve for Unknown Force - Jack Force through desired translation distance (you might first have to solve for desired translational distance based on desired rotation angle).

[JDMather]

@AnthonyHarris90842 

Split the model in half to simplify the set up of constraints (Joints) and solve for only one side.

This will give the force for one cylinder which is equivalent in the second cylinder.

Otherwise you will likely be plagued with overdefined conditions in the Joint solver.

You will have a Revolution Joint for the Rotation DoF.

Constrain the Cylinder for Revolution Joint at one end and Point-Line or Point-Point at the other end.

Constrain the Cylinder Piston such that it has a single Translational DoF.

Solve for Unknown Force - Jack Force through desired translation distance (you might first have to solve for desired translational distance based on desired rotation angle).

[AnthonyHarris90842]

Thank you so much.
I'll try that.

[AnthonyHarris90842]

Thank you so much.
I'll try that.

[JDMather]

@AnthonyHarris90842 

If you still have trouble figuring it out - post a screen shot of your Joints list.

[JDMather]

@AnthonyHarris90842 

If you still have trouble figuring it out - post a screen shot of your Joints list.