Discussion Groups

## Simulation Mechanical and Multiphysics

Contributor
Posts: 14
Registered: ‎08-13-2011

714 Views, 17 Replies
08-22-2011 02:14 AM

can any body sort it out.

whenever i perform steady state heat transfer, result shows the body initial temperature 0 degree C even i wrote 20 degree C in default nodal temperature, i'm applying 50 degree see to my model and it contain 8 parts; 6 solid and 2 liquid.

Employee
Posts: 492
Registered: ‎03-25-2010

# Re: steady state heat transfer

08-22-2011 05:16 AM in reply to: chikhlas

Hello,

I do not have any explanation. What version are you using? In particular, what loads do you have applied to your model?

More importantly, do the results look to be correct? Technically, the initial temperature in a steady state solution does not affect the results. You can put in an initial temperature of 1E6 degrees, and the results will be the same after an infinite amount of time (steady state). Providing a reasonable initial temperature can be beneficial when temperature dependent material properties or surface radiation are included in the analysis; the initial temperature provides a "seed" value to speed up the convergence iterations.

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)
Contributor
Posts: 14
Registered: ‎08-13-2011

# Re: steady state heat transfer

08-22-2011 06:11 AM in reply to: INACTIVE_AstroJohn

i'm using autodesk simulation 2012, i apply a load ''applied temperature '' to a surface 50degree then i also mention default nodal temperature 20 degree in analysis parameter but still my result starts from zero degree,

Contributor
Posts: 14
Registered: ‎08-13-2011

# Re: steady state heat transfer

08-22-2011 06:29 AM in reply to: chikhlas

i wish the top plate should be at surrounding temperature 20 degree, i am applying 50 degree at the bottom, in the middle the chamber contain air. there will be conduction in the bottom plate then heat will transfer to the air enclosed in chamber. i generated fluid inside the chamber. i'm also bit confused either i'll apply convection load to solid surface which interact with fluid or i'll apply convection load to fluid surface which is adjacent to hot solid temperature. due to convection the temperature of air will increase then it will interact with top plate which should be at 20 degree and heat will be transfer again. i hope it can give an understanding.

there is also a piston inside the chamber but i got a reply from autodesk advisor that it will not work, coupling is only for fluid flow and heat transfer in multiphysics

Employee
Posts: 492
Registered: ‎03-25-2010

# Re: steady state heat transfer

08-22-2011 08:13 AM in reply to: chikhlas

Thanks for the image. It helps to explain what is going on. Here are a number of things for you to consider.

1. I think you are confusing the initial temperature ("Setup > Thermal Loads > Initial Temperature" and "Setup > Model Setup > Parameters > Options", both which set the temperature AT TIME 0) and the calculated temperature which is shown in your image. As mentioned before, the calculated temperature ("Results Contours > Temperature > Calculated Temperature") is at time infinity in a steady state analysis, and the calculated temperatures are not affected by the initial temperature. For what it is worth, the "Results Contours > Temperature > Initial Temperature" should show the model shaded with temperatures of 20 degrees.
2. Did you get any warning messages from the analysis? If so, go to the Report tab and choose to view the log file or summary file to find the warnings.
3. Based on your written description, the model has a source of heat (50 degree applied temperature, "Setup > Thermal Loads > Controlled Temperature") but no means of removing the heat. Therefore, a steady state thermal analysis should have heated the model to a uniform temperature. Sometimes the analysis does this; other times, it gives a warning that the solution is unstable, or missing a load, or there may not be a unigue solution, or something like that, but it gives a result which is not accurate. (Garbage in, garbage out.)
4. For the "surrounding temperature of 20 degrees" that you wrote about for the top, you should apply a load that represents this. A convection load ("Setup > Thermal Loads > Convectin") is usually the type of load that simulates this the best since you know what surrounds the model (air?), the orientation of the surface (horizontal?), and the approximate temperature of the part (less than 50 degrees?), so the convection coefficient can be calculated.
5. If doing just a "Thermal > Steady-State Heat Transfer" analysis, you do not want to apply a convection load to the inside chamber. Most loads in thermal analysis represent a transfer of heat from/to the model and the environment. That is, it adds or removes heat from the model. In your situation, heat is being transferred from one part to another; it is not escaping the model.
6. The best that can be done with the internal air pocket in Thermal analysis is to let the internal air conduct the heat across the air chamber. Depending on what the analysis represents, this may be sufficient, but the thermal conductivity of the air probably needs to be increased to represent the air currents that exist in reality.
7. I do not know if the educational version includes the analysis type "Multiphysics > Steady Coupled" or "Multiphysics > Transient Coupled" or not. Those two analysis types will calculate the air currents (convective loads) that occur in the internal  air chamber.

Hope this gives you some ideas. Don't hesitate to ask for additional ideas :-)

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)
Contributor
Posts: 14
Registered: ‎08-13-2011

# Re: steady state heat transfer

08-22-2011 08:33 AM in reply to: INACTIVE_AstroJohn

thanks i'm trying these, can u also tell me about contact type, there is an option in tree at the bottom e.g bonded, surface contact. Do i need to choose any of these for my model or i should go on by default bonded. as i told you these is a piston inside the chamber, for fluid flow as well so the air stream should not intersect the piston.

Is it possible to move the piston in multiphysics, thermal or fluid flow analysis due to air pressure?

i may have few more questions but first i'm going to try your suggestions.

onething more when i get result and try to generate graphs, my software stuck and some time it corrupts the file.

when i try to make streamlines in fluid, it always stuck for a while.

Employee
Posts: 259
Registered: ‎06-14-2010

# Re: steady state heat transfer

08-24-2011 07:38 AM in reply to: chikhlas

(there is an option in tree at the bottom e.g bonded, surface contact. )

A: Contact is talking about the surface relations between parts.

-Welded: same as bonded, two parts are fully connected like a single part

-Free/No contact: two parts have no connections

-Surface contact:  can define contact “total thermal resistance” between two surfaces

(Is it possible to move the piston in Multiphysics, thermal or fluid flow analysis due to air pressure?)

A: We don’t support two-way FSI (fluid-solid interaction) in Multiphysics, however, one work around is that user can do multiple steps using one way FSI.

(when i try to make streamlines in fluid, it always stuck for a while.)

A: Using finer mesh will help, and it needs enhancement in our POST as well.

﻿

Jianhui Xie, Ph.D
Principal Engineer
MFG-Digital Simulation
Employee
Posts: 259
Registered: ‎06-14-2010

# Re: steady state heat transfer

08-24-2011 07:51 AM in reply to: Joey.X

(i told you these is a piston inside the chamber, for fluid flow as well so the air stream should not intersect the piston.)

A: The thermal contact between air stream and solid part usually set to be bonded since the fluid makes very good contact with solid surface, however, if there are coating on the solid part surface but the coating layer is not modeled, the fluid-solid contact should be considered. In comparison, the roughness in solid to solid surface makes imperfect contact, thus needs to provide thermal resistance between parts.

﻿

Jianhui Xie, Ph.D
Principal Engineer
MFG-Digital Simulation
Contributor
Posts: 14
Registered: ‎08-13-2011

# Re: steady state heat transfer

08-31-2011 04:14 PM in reply to: Joey.X

i'm still facing problem, as i told you earlier. the default nodal temperature in my autodesk does not work, my results shows temperature zero degree, although i set default nodal 20 degree.

applied temperature 50 degree

now i removed fluid convection as well

kindly see below ds.lgs

Autodesk (R) Simulation Steady-State Heat Transfer
Version 2012.01.00.0017-W64/X64 15-Jun-2011

**** Memory Dynamically Allocated =   1928538 KB

----------------------------------------------------------------
DATE: AUGUST 31, 2011
TIME: 11:27 PM

PROGRAM VERSION: 201201000017
ALG.DLL VERSION: 201201000017
AlgConfig.DLL VERSION: 201201000017
Agsdb_AR.DLL VERSION: 201201000017
AMGSolve.DLL VERSION: 201201000017
AlgSolve.DLL VERSION: 201201000017
----------------------------------------------------------------

Options executed are:
NOMIN
SUPSTR

processing ...

**** Model Unit System Settings:
--------------------------------------------
Unit System              : Custom
Force                    : N
Length                   : mm
Time                     : s
Temperature (Absolute)   : deg C (K)
Thermal Energy           : J
Voltage                  : V
Current                  : A
Electrical Resistance    : ohm
Mass                     : N*s^2/mm
--------------------------------------------

**** OPENING TEMPORARY FILES
NDYN = 10
**** BEGIN NODAL DATA INPUT
Node     number=    256493
Equation number=    256493
**** END   NODAL DATA INPUT
**** HARD DISK FILE SIZE INFORMATION FOR PROCESSOR:

Available hard disk space on C drive = 115292.543 megabytes

**** BEGIN NODAL HEAT FLOW INPUT
**** END   NODAL HEAT FLOW INPUT
**** BEGIN TYPE-39 DATA INPUT
3158 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
8795 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
3270 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
106516 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
8350 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
8876 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
43396 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
11279 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TEMPERATURE ELEMENT DATA INPUT
573 ELEMENTS ...
**** END   TEMPERATURE ELEMENT DATA INPUT
**** Invoking BCSLIB-EXT Sparse Solver ...
**** End solution

**** BEGIN TYPE-39 DATA INPUT
3158 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
8795 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
3270 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
106516 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
8350 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
8876 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
43396 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TYPE-39 DATA INPUT
11279 ELEMENTS ...
**** END   TYPE-39 DATA INPUT
**** BEGIN TEMPERATURE ELEMENT DATA INPUT
573 ELEMENTS ...
**** END   TEMPERATURE ELEMENT DATA INPUT
**** BEGIN TYPE-39 HEAT FLUX CALCULATIONS
3158 ELEMENTS ...
**** END   TYPE-39 HEAT FLUX CALCULATIONS
**** BEGIN TYPE-39 HEAT FLUX CALCULATIONS
8795 ELEMENTS ...
**** END   TYPE-39 HEAT FLUX CALCULATIONS
**** BEGIN TYPE-39 HEAT FLUX CALCULATIONS
3270 ELEMENTS ...
**** END   TYPE-39 HEAT FLUX CALCULATIONS
**** BEGIN TYPE-39 HEAT FLUX CALCULATIONS
106516 ELEMENTS ...
**** END   TYPE-39 HEAT FLUX CALCULATIONS
**** BEGIN TYPE-39 HEAT FLUX CALCULATIONS
8350 ELEMENTS ...
**** END   TYPE-39 HEAT FLUX CALCULATIONS
**** BEGIN TYPE-39 HEAT FLUX CALCULATIONS
8876 ELEMENTS ...
**** END   TYPE-39 HEAT FLUX CALCULATIONS
**** BEGIN TYPE-39 HEAT FLUX CALCULATIONS
43396 ELEMENTS ...
**** END   TYPE-39 HEAT FLUX CALCULATIONS
**** BEGIN TYPE-39 HEAT FLUX CALCULATIONS
11279 ELEMENTS ...
**** END   TYPE-39 HEAT FLUX CALCULATIONS

**** Table for actual hard disk space used:
ds.t7  =    2003.859 kilobytes
ds.t8  =   11066.250 kilobytes
ds.t9  =      41.250 kilobytes
ds.t10 =      41.250 kilobytes
ds.t11 =  403086.422 kilobytes
ds.t12 =       0.000 kilobytes
ds.t13 =      41.250 kilobytes
ds.t14 =       0.562 kilobytes
ds.t15 =      41.250 kilobytes
ds.t18 =      41.250 kilobytes
ds.t65 =    2003.859 kilobytes
ds.t67 =  102639.617 kilobytes

total temporary disk storage (megabytes) = 508.796

**** BEGIN DELETING TEMPORARY FILES
**** TEMPORARY FILES DELETED
**** END OF SUCCESSFUL EXECUTION

Processing completed for model C:\Users\Ikhlas\Documents\Final Project\acad\789.ds_data\11\ds

ds.l10   =       3.885 kilobytes
ds.to    =   12023.156 kilobytes
ds.hfo   =    6807.828 kilobytes

Total actual hard disk space used    =    533.837 megabytes

Total elapsed time                   =     25.635 minutes

﻿

Employee
Posts: 492
Registered: ‎03-25-2010

# Re: steady state heat transfer

09-01-2011 05:45 AM in reply to: chikhlas

Hi chikhlas,

Either your description is inaccurate, or your model is setup incorrectly. I believe you are confusing the default nodal temperature with the minimum calculated temperature. As I tried to explain before, the default temperature that you entered under the Analysis Parameters dialog has absolutely no affect on the calculated temperatures in steady-state heat transfer. So, the minimum calculated temperature of 0 degrees shown in your image "model.jpg" (attached to the 8-22-2011 post) is not related to the 20 degrees default temperature you applied. In the results environment, go to "Results Contours > Temperature > Initial Temperature" on the ribbon; you will see the entire model is 20 degrees.

For the calculated temperature results, one possibility is that you have some other load applied to the model that is forcing the temperature to be 0. Another possibility is the solution is wrong because you only have 1 load applied (the 50 degrees on the bottom surface). Another possibility is that the top plate is not connected to the lower stuff, so heat cannot transfer from below to the top. (I am basing this last possibility on the model.jpg image you posted.)

I think that neither the log file (ds.Lgs) or summary file (ds.L10) contains the data necessary to determine the problem. So you will either need to send the model to technical support, or we can try to get the file small enough to post to this forum (the limit is 2 or 3 MB if I remember correctly), or post it to a common site like DropBox (www.dropbox.com) or You Send It (www.yousendit.com). Here's what you should do to get the archive as small as possible. Follow each step carefully so that you do not accidentally destroy your real model!

1. Save the model to a new name. [With the model open, click the "S" application button, then "Save As". Give the model a new name like "initial temperature".]
2. Delete all of the design scenarios except for 11 in the new model. [This is the design scenario you showed in your log file. Click the first design scenario in the browser, hold the shift key, and click the last design scenario before 11. Right-click on one of the selections and choose "Delete".]
3. Use Windows Explorer (My Computer) to navigate to the folder containing the model. Then navigate to the "initial temperature.ds_data\11\ds.mod" folder.
4. Delete all of the files EXCEPT for model.cdx, model.dbf, and model.ftp. [Depending on your Windows Explorer, the extensions may not be shown. But there are only three files named "model", so delete everything else.]
5. Archive the model only. [In Autodesk Simulation, click the "S" application button, then "Archive > Create > Model Only".]
6. Post the archive file ("initial temperature.ach") somewhere. Attach to this forum if it is smaller than whatever the limit is.

Thanks.

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)