Learning Forum

[salvorusso946]

Hi everyone,

 

I'm running an explicit thermomechanical analysis in LS-DYNA and I'm trying to speed up the simulation using time scaling, by applying an appropriate TSF (Thermal Speedup Factor).

 

To evaluate the impact, I ran two simulations:

 

1. One with time scaling, using a TSF to modify thermal properties accordingly.

 

 

2. One without time scaling, i.e., standard time and no TSF applied.

 

 

 

Surprisingly, the simulation with TSF results in higher temperatures compared to the one without time scaling. I was expecting both simulations to show the same thermal behavior (just occurring over different time scales), but it seems that the temperature evolution is affected in a nontrivial way.

 

Has anyone encountered this issue before? I'm trying to understand:

 

Is this temperature increase a known side effect of using TSF?

 

Are there specific best practices or precautions when using TSF (e.g., adjusting boundary conditions, time step, or coupling settings)?

 

Could the discrepancy be related to how I’m scaling the thermal material properties or defining the thermomechanical coupling?

 

 

Any guidance or shared experience would be greatly appreciated. Thanks in advance!

[Nanda]

Hello user,

What's the TSSFAC value you used? Too much added mass tends to affect the results. I will let someone else comment on the temperature part. I'm sharing the general guidelines of using mass scaling:

LSDYNA Mass scaling

Is it bad practice to reduce safety factor instead of Mass Scaling?

Regards,

Nanda.

Introducing Ansys Assistant a 24/7 AI-powered virtual assistant for active Ansys academic customers.

Have a look at our public help documentation website: Ansys Help

For more exciting courses and certifications, hit this link: Ansys Innovation Courses | ANSYS Innovation Space

Guidelines for Posting on Ansys Learning Forum

[salvorusso946]

Thank you very much for the helpful response!

 

In my case, I used TSSFAC = 0.4. Based on your suggestion, I’ll try reducing the mass scaling and see how it affects the results. I’ll run some tests and compare again with the reference (non-scaled) simulation.

 

In the meantime, I’m still hoping to hear from others as well — especially regarding possible thermal effects or side-effects introduced by using TSF. Any further clarification on that side would be greatly appreciated.

 

Thanks again!

 

[Dennis Chen]

can you help me understand a little bit about why you are doing a thermal-mechanical coupled analysis explicitly?    Is it something specific to your problem that this is required?   In most cases, I would prefer to do this implicitly given the time scale of the problem (for a structure to reach steady state thermal condition, it takes a while)

in the docs, the use application they described is for stamping which just suggests the same thermal scaling as your mass scaling.    

This slide is really good and it described when to scale thermal velocity when mechanical velocity is scaled - https://ftp.lstc.com/anonymous/outgoing/support/FAQ_docs/heat_transfer_class.pdf

[salvorusso946]

Thank you very much for your detailed and helpful replies! I truly appreciate the time and insights you have shared.

 

I have chosen to use an explicit integration method for my thermomechanical coupled analysis mainly due to the strong nonlinearity present in my process. This makes the explicit approach more suitable despite its challenges with time scaling.

 

Regarding the document you kindly cited, I was already familiar with it and have closely followed the recommendations and guidelines contained in that very useful file. It has been instrumental in setting up and understanding the scaling process in my simulations.

 

If possible, I would be interested to know:

 

What practical advice would you give to ensure thermal behavior remains accurate when applying time scaling in explicit coupled analyses?

 

Are there any specific boundary condition or coupling parameter adjustments that you consider critical to avoid side effects such as unexpected temperature increases?

 

How do you typically validate or benchmark time-scaled thermomechanical explicit simulations to be confident of their physical accuracy?

 

Thanks again for your support and advice!

[salvorusso946]

I would like to specify that the analysis I am conducting uses an explicit integration scheme for the mechanical part, while the thermal part is solved using an implicit method. This approach helps me better handle the nonlinear mechanical behavior alongside stable thermal calculations.

 

Please let me know if you have any suggestions or insights regarding this mixed integration strategy. Thank you again for your valuable support!