Discussion Groups

## Simulation Mechanical and Multiphysics

Contributor
Posts: 16
Registered: ‎12-08-2011

# time step reduction even when the iteration converge !!!!!!!!

317 Views, 7 Replies
04-04-2012 10:59 AM

** Time step reduction due to: concrete material

i'm modeling a reinforced concrete, i keep getting this message, "
** Time step reduction due to: concrete material "

from HELP:
"
NoteThere are several criteria that must be met for the time step to be converged. The letter c after the Residual indicates that the equilibrium loop has converged. After printing the c, other conditions are checked for convergence. Thus, the log file may imply that the solution has converged and then reduce the time step! When this occurs, the log file will display a message such as Time step reduction due to: contact penetration to indicate the reason for the lack of convergence.
"
can anyone point out the problem ?! why did it reduce time step till diveregance even though the iteration converge, and what are the other parameters and criteria that the analysis check upon ?!

Please use plain text.
Employee
Posts: 492
Registered: ‎03-25-2010

# Re: time step reduction even when the iteration converge !!!!!!!!

04-10-2012 11:11 AM in reply to: 80DAVE

Hi Dave,

My guess would be that the load is too large for the strength of the concrete. Therefore, check these items:

(1) Use the load curves to ramp the applied loads from 0 to the maximum load over a number of time steps.

(2) Check the concrete material properties to make sure nothing was left at 0 strength.

Otherwise, change all of the concrete parts to the Isotropic material model, choose the concrete material from the library, and run the simulation. If this converges, then the problem is with the concrete parts. If it still does not converge, then the problem is with something else. For example, is there surface to surface contact in the simulation?

Sincerely,
John Holtz, P.E.
Senior User Experience Designer, Simulation
Autodesk, Inc.

Current version of Mechanical & Multiphysics: 2013 SP1 (2013.01.00.0012 28-Jun-2012)
Please use plain text.
Contributor
Posts: 16
Registered: ‎12-08-2011

# Re: time step reduction even when the iteration converge !!!!!!!!

04-10-2012 01:54 PM in reply to: INACTIVE_AstroJohn

first of all, thank you for replying, i was waiting so long .!

regarding your notes, i'm

using a standard load curve, starts from zero, till 20 seconds .! linear curve.

i'm using a default contact (bonding) between steel and concrete nodes.

about changing the simulation to istotropic linear, it worked out !!!!!!!!! converged and done .!

i'm guessing it's from the strength parameters ?! i'm defining it according to a laboratory results .!

uniaxial tensile (2385), uniaxial compression (33072), biaxial compressive (39686.4), all in KN/m2, retension (1)

the only thing i'm not defining is the tension stiffening (none), do you think it has something with divergence ?!

BTW, the model initial, analyze till a certain point (almost before uniaxial tensile), then it collapsed and diverge .!

Please use plain text.
Employee
Posts: 492
Registered: ‎03-25-2010

# Re: time step reduction even when the iteration converge !!!!!!!!

04-11-2012 09:12 AM in reply to: 80DAVE

Hi,

Your issue could be easy to solve, or it could become quite involved. No one will know for sure until you provide an archive of your model. See the post Create, Post, or Provide an Archive of your model.

I suggest that you contact Technical Support and have them handle it. Here's how to do this:

http://subscription.autodesk.com/

You will need to log on using your Autodesk user id and password. From within the subscription center, you will be able to view past support issues, enter comments and open new support cases. If you are not listed as a contact under the serial number, then you will need to contact the account contact so that he or she can log on to the subscription center and add you as a contact under this serial number.

Alternatively, you can call 1-866-487-8680, then choose option "6", then option "4". If we are all on the phone, then it will be best to log an issue on the subscription center. The second best option would be to leave a message. The least favorable option may be to remain on hold.

Sincerely,
John Holtz, P.E.
Senior User Experience Designer, Simulation
Autodesk, Inc.

Current version of Mechanical & Multiphysics: 2013 SP1 (2013.01.00.0012 28-Jun-2012)
Please use plain text.
Contributor
Posts: 16
Registered: ‎12-08-2011

# Re: time step reduction even when the iteration converge !!!!!!!!

04-12-2012 03:58 AM in reply to: INACTIVE_AstroJohn

i've tired to log in http://subscription.autodesk.com/, but something went wrong, server thing is out of time mmmmm, or something like that, anyway, i'm including the model (archieve) in the attachement, discrete model for quarter reinforced concrete column.

if you would examin it, and see why it diverge, i'd be more than grateful.

regards.

Please use plain text.
Contributor
Posts: 16
Registered: ‎12-08-2011

# Re: time step reduction even when the iteration converge !!!!!!!!

04-12-2012 03:57 PM in reply to: 80DAVE

dear John,

i've called the number you gave me, all i got is an answer machine .

best regards .!

Please use plain text.
Contributor
Posts: 16
Registered: ‎12-08-2011

# Re: time step reduction even when the iteration converge !!!!!!!!

04-20-2012 09:52 AM in reply to: 80DAVE

still waiting, people .!

Please use plain text.
Employee
Posts: 188
Registered: ‎10-04-2010

# Re: time step reduction even when the iteration converge !!!!!!!!

04-30-2012 06:59 AM in reply to: 80DAVE

In the nonlinear iterative process, the Yield/Failure check for concrete material should be done following each iteration and step.  Thus, even the equilibrium is convergent (shown with "C" or "Cr" following in log file), further Yield/Failure check are still needed for this iteration.  The Yield/Failure check for concrete material is also another iterative process, which also has a convergent check.  If the Yield/Failure check concrete material cannot converge, Autodesk Simulation will cut time step even the equilibrium is convergent.

Acutually, for a convergent step, there are some other reasonss, which will result in time reduction,

such as "element Jacobian error in shape function", "element Jacobian error in stress calcualtion", "creep calculation",

"numerical checking in Mooney-Rivlin model", "shell element calculation", "contact penetration", etc.

Hope this can help.

Shoubing

Please use plain text.