Facing an error as Static Solver could not converge during quasi-static simulation

Sudip_Chavan

Trying to Run a MBD model with quasi-static simulation showing Error: Static solver could not converge [1]

Showing Possible Cause as per following attachment below 

For Possible Cause 1 & 2: I tried iterating simulation settings, deactivated all user output requests and only trying to generate h3d file, still getting a same error with all possible combinations of simulation settings. 

For Possible Cause 3: I checked and verified that there is not any Non-physical inertia properties present in the model

Not able to understand what can be changed to get rid of this error. 

GTT Adam

Hi Sudip,

Unfortunately from the image 'MBD Error.png' it is not really possible to find out what the direct cause is. Your model is not initializing, which means there is a major error which does not let the integrator proceed past time=0. 

More information can be gathered by reviewing the messages panel at the bottom of the MotionView interface. This may show errors relating to missing entities or external files that can no longer be found.

As well, you can enable DEBUG mode for the LOG file, which will give more information about residual error in joints and bodies prior to lockup.

 

Hope this helps!

Adam Reid

Praful

Hello Sudip -

Could you please check

1. if there are any un-constrained bodies like dummy bodies ? A few ways to verify

• • Turn on the Suspect data in the MotionView preferences. (If you are using the Classic interface, it will be under Tools > Options menu)

• • Run a dummy transient run for a small time period. In HyperView, turn on all markers (Systems) and observe if one or more markers are falling off.
• There could be bodies that are not fully constrained, like a rotating body that might not be able to get static equilibrium or the equilibrium position may be too far away from the initial position.

A transient run can reveal these situations.

Hope this helps.

Regards

Praful

 

Sudip_Chavan

Adam Reid_21142 said:

Hi Sudip,

Unfortunately from the image 'MBD Error.png' it is not really possible to find out what the direct cause is. Your model is not initializing, which means there is a major error which does not let the integrator proceed past time=0. 

More information can be gathered by reviewing the messages panel at the bottom of the MotionView interface. This may show errors relating to missing entities or external files that can no longer be found.

As well, you can enable DEBUG mode for the LOG file, which will give more information about residual error in joints and bodies prior to lockup.

 

Hope this helps!

Adam Reid

Hi Adam,

Thanks for the info, I will check DEBUG mode and will try to gather more info to solve this error.

Actually I simulated my model in static simulation with output settings as "Generate animation at each iteration" turned on and found out that my model is not behaving correctly in static condition as it should be. 

Sudip_Chavan

Praful Prabhu_20784 said:

Hello Sudip -

Could you please check

1. if there are any un-constrained bodies like dummy bodies ? A few ways to verify

• • Turn on the Suspect data in the MotionView preferences. (If you are using the Classic interface, it will be under Tools > Options menu)

• • Run a dummy transient run for a small time period. In HyperView, turn on all markers (Systems) and observe if one or more markers are falling off.
• There could be bodies that are not fully constrained, like a rotating body that might not be able to get static equilibrium or the equilibrium position may be too far away from the initial position.

A transient run can reveal these situations.

Hope this helps.

Regards

Praful

 

Hello Praful,

Thanks for the info!

1. I updated the preferences and checked the info in check model window, there is no any free body list.

2. I simulated my model in transient run and found out the tires are steering in static itself which should not be the case for ride analysis event. 

Hopefully there are not any markers falling down in animation but the wheels are steering in one direction in initial time step itself from this I concluded that the rotating body (wheels) are not able to get static equilibrium, I will try to update my modelling parameters and will let you know about the results.

 

Thanks a lot!

Regards

Sudip Chavan

Sudip_Chavan

Hello Everyone!

I updated my various modeling parameters but still not able to achieve equilibrium condition in static.

I am attaching the latest log file of my model to this post.

Can anyone please suggest me some solutions or some other methods to cross verify the model? 

 

Thanks and Regards

Sudip Chavan

Praful

Sudip Chavan_20509 said:

Hello Everyone!

I updated my various modeling parameters but still not able to achieve equilibrium condition in static.

I am attaching the latest log file of my model to this post.

Can anyone please suggest me some solutions or some other methods to cross verify the model? 

 

Thanks and Regards

Sudip Chavan

Sudip - I see this error in the log.

WARNING: Failure in parsing "(ATAN(DZ(30301,30306,30101010) /DY(30301,30306,30101010)))*RTOD" for Request with id = 70000037!
          Marker [id=30301] not found in expr "DY"!]
          Execution continues but this expr is set to identical zero!

Looks  like these are similar errors in the expression like I showed in the solver array (different thread)

Of course, this may not be related to the static convergence. Is this model different than the one you shared earlier ? Because that model ran fine as soon I fixed the solver array marker ids.

regards

Praful

GTT Adam

Sudip Chavan_20509 said:

Hello Everyone!

I updated my various modeling parameters but still not able to achieve equilibrium condition in static.

I am attaching the latest log file of my model to this post.

Can anyone please suggest me some solutions or some other methods to cross verify the model? 

 

Thanks and Regards

Sudip Chavan

Hi Sudip,

Have you tried running the model as a different analysis type? If possible, I would recommend trying a Transient analysis so that you skip the Static simulation. If it initializes and you can review an animation, it may help to highlight a modelling error that is preventing the Static simulation from ever levelling out. A simple example would be of a falling ball, where equilibrium can not be found since it is falling forever. A Transient animation may highlight that.

 

Hope this helps!

Adam Reid

Sudip_Chavan

Praful Prabhu_20784 said:

Sudip - I see this error in the log.

WARNING: Failure in parsing "(ATAN(DZ(30301,30306,30101010) /DY(30301,30306,30101010)))*RTOD" for Request with id = 70000037!
          Marker [id=30301] not found in expr "DY"!]
          Execution continues but this expr is set to identical zero!

Looks  like these are similar errors in the expression like I showed in the solver array (different thread)

Of course, this may not be related to the static convergence. Is this model different than the one you shared earlier ? Because that model ran fine as soon I fixed the solver array marker ids.

regards

Praful

Hello sir, 

It is totally different model from which I shared with you. This model is completely built from scratch and suspension system is completely different from all the systems available in the template models. 

I deactivated all the outputs involved in the model and only trying to achieve correct rigid body model that's why I believe the solver array should not be the issue, still I cross verified & updated all the solver arrays as told by you.

 

Regards 

Sudip Chavan

Sudip_Chavan

Adam Reid_21142 said:

Hi Sudip,

Have you tried running the model as a different analysis type? If possible, I would recommend trying a Transient analysis so that you skip the Static simulation. If it initializes and you can review an animation, it may help to highlight a modelling error that is preventing the Static simulation from ever levelling out. A simple example would be of a falling ball, where equilibrium can not be found since it is falling forever. A Transient animation may highlight that.

 

Hope this helps!

Adam Reid

Hello sir,

Yes, I ran transient analysis successfully. In transient file results the system just falls down by some displacement for 0.1 sec and after that both wheels steers by some degree for till 10 seconds that's it. 

No any system falls down or no any system disconnects from its specific body only connections of drop link with stabbar by bushings are breaking out and droplinks are just waving in air due to connection with ground body at other end. Wheel jacks are not moving according to motion defined to it. 

I can show you my file on a online meeting.

Is it possible for you to connect with me online over any platform Gmeet/Microsoft Teams/Webex/Zoom whatever suits you?

Regards

Sudip Chavan

Sudip_Chavan

Praful Prabhu_20784 said:

Sudip - I see this error in the log.

WARNING: Failure in parsing "(ATAN(DZ(30301,30306,30101010) /DY(30301,30306,30101010)))*RTOD" for Request with id = 70000037!
          Marker [id=30301] not found in expr "DY"!]
          Execution continues but this expr is set to identical zero!

Looks  like these are similar errors in the expression like I showed in the solver array (different thread)

Of course, this may not be related to the static convergence. Is this model different than the one you shared earlier ? Because that model ran fine as soon I fixed the solver array marker ids.

regards

Praful

Hello sir,

I can show you my model on a online meeting.

Is it possible for you to connect with me online over any platform Gmeet/Microsoft Teams/Webex/Zoom whatever suits you?

 

Regards 

Sudip Chavan

Praful

Sudip Chavan_20509 said:

Hello sir,

Yes, I ran transient analysis successfully. In transient file results the system just falls down by some displacement for 0.1 sec and after that both wheels steers by some degree for till 10 seconds that's it. 

No any system falls down or no any system disconnects from its specific body only connections of drop link with stabbar by bushings are breaking out and droplinks are just waving in air due to connection with ground body at other end. Wheel jacks are not moving according to motion defined to it. 

I can show you my file on a online meeting.

Is it possible for you to connect with me online over any platform Gmeet/Microsoft Teams/Webex/Zoom whatever suits you?

Regards

Sudip Chavan

Hi Sudip

In the animation in HyperView, turn on all the markers and see if there is anything unusual.

Review part by part joint configurations and see of there is any possibility of unconstrained d.of. e.g steering or tierod having ball joints at the ends will have a rotation d.o.f still unconstrained.

I can certainly join on a Teams call on Monday. Which part of the world are you located ?

regards

Praful

 

Sudip_Chavan

Praful Prabhu_20784 said:

Hi Sudip

In the animation in HyperView, turn on all the markers and see if there is anything unusual.

Review part by part joint configurations and see of there is any possibility of unconstrained d.of. e.g steering or tierod having ball joints at the ends will have a rotation d.o.f still unconstrained.

I can certainly join on a Teams call on Monday. Which part of the world are you located ?

regards

Praful

 

Hello Sir,

I'm located in Pune, India.

Can we have Teams call at 2.30 PM IST Today?

Please confirm on my mail: sudip.c@ekamobility.com, I will revert you with Teams Invite link.

 

Thanks and Regards

Sudip Chavan