Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Gypsum - Steel thermal analysis 1

Status
Not open for further replies.

Juliet27

Civil/Environmental
Jun 24, 2005
10
0
0
GB
Hello Ansys Users,

My name is Juliet and im a new user of Ansys.
The purpose of my post is because i would like to ask you a question.

I have to do a conduction analysis of a 2D element consisted of two layers.
The first one is gypsum with specific thermal properties and the second is steel.
When im assignning temp on the gypsum line and doing the analysis i can see that temperature doesnt conducted to the steel layer (im using gupsum as insulation).

I have used the GLUE command to glue the areas but the result is the same.

When im performing thermal analysis only to the gypsum leayer i can see that the outer line of the layer has approx 400 degrees of celsius.
When im performing the same procedure with steel on the outer place of the gypsum no temperature is conducted to the steel.

Please i would appreciate if you could help me

regards
 
Replies continue below

Recommended for you

If these are two separate entities then you need to create (thermal) contact elements between them. The contact will allow transfer of the thermal loads as well as allowing you to specify the properties of the thermal interaction (conduction etc.). If you haven't used contact before you will need to read the excellent section in the help file:

Structural Guide> Chapter 11. Contact>11.4.10. Modeling Thermal Contact

or do a search on "thermal contact" and click on the "11.4 Performing a Surface-to-surface contact..." option.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com
 
thank you a lot,

so may i ask you something,

do you think i dont have to glue but only to use the contact pair command?

thank you for your help
 
Correct. Don't glue the areas, use contact elements (a "contact" and "target" pair - 170 and 174 for example).

Cheers,

-- drej --


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com
 
just to make clear,

imagine to rectangle areas (1)(2)
The first one is gypsum and the second one is steel.
Im meshing the areas and then its says pick target pick pair.
Im a liitle bit confused.
Because i was using the GLUE command for areas now do i have to use areas again?
Which element is my target and which is pair (the help doesnt help me)

Im assigning thermal on gypsum and then i want to see the thermal distribution to the steel.


Thank you a lot Drej
 
Do you think that Glue command creates the problem
Im using a PLANE55 element and im perfoming a thermal conduction analysis.

im a bit confused
 
I'm confused as well.

Forget the glue thing, it's irrelevant I'm sure (I can't visualise what/why you're trying to do by glueing the two areas). Before we start going round in circles, explain to me your problem, including why you chose plane55 and why you split up the steel and the gypsum into two areas and tried to glue them. Is it just a layer of steel on a layer of gypsum? Why are you using plane55 for this? What are trying to achieve? Is there an example of the steel/gypsum assembly anywhere to help me visualise the problem?


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com
 
Im perfoming a thermal transient analysis.

i have two thin elements in similar dimensions:
Height 330mm and thickness 10mm.
One is made by gypsum (1) and the the other (2) is made by steel.

Like the following pic
.........
. . .
. . .
. 1 . 2 .
. . .
. . .
.........

Im applying temp on the outside line of the gypsum and i whant to investigate the temperature on the steel at different locations.
Im using thermal mass: Plane55 2D analysis

The problem is that i was thinking to use the GLUE command to glue the two areas but the problem is that tempeature distribution can be seen only to the gypsum and not to the steel (approx 20 degrees is the temp on the steel evethough the temp is about 1000 degrees).

As you told mentioned me before and im glad for that you told me to contact the elements (a "contact" and "target" pair).
By reading the Help extensively i couldnt underastand a lot.
I can understand what is target and what is contact and what i have to pick lines or areas?
How i can use this command.


Thank you a lot Drej.

 
It's still not clear to me what you're doing. Are you taking a cross section through the steel/gypsum:

-------------------------
| | |
| S | G |
| T | Y |
| E | P |
| E | S |
| L | U |
| | M |
| | |
| | |
-----------------------

If so you just need to create two areas. Assign each area the appropriate material/type/real value (see AATT command)and then mesh each area. No contact needed, no glue command needed. Although this sounds suspiciously like what you've done already...


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com
 
Yes Grej thats right is a cross section of two elements.

But the problem is when im assigning temp on gypsum there is no temp distribution on steel, i would like to mention that gyspum is not such a great insulation material.

To test if gypsum is so good material i ve just made only the gypsum element ( cross section ) and i pefomed thermal analysis and i saw that the temperature to the other side is approx the half (400 degrees of celsius).

SO i was thinking that half of this temp is going to be conducted to the steel but the temp on the steel side when are glued is only 14 degress.

I think that this is sounds strange.

You mentioned to me to use the pair command but i still im bit condused how that it works, i ve read the help a lot of times but i dont undersatnd which target and which is the pair and also i have to pick areas or lines?

Sorry for the ammount of my posts but please it would be helpful if you could help me.


Thank you a lot Drej.
 
Read my previous post. Forget contact and the glue command you don't need them. My advice is to start from the beginning.

1/ Generate two areas side-by-side as I've drawn above - area 1 will be gypsum, area 2 steel
2/ generate the thermal properties for steel in material 1 (MP,#,1), material 2 (MP,#,2) will be the gypsum
2/ assign area 1 the properties of steel (AATT command), and area 2 the gypsum
3/ mesh each area
4/ apply your boundary conditions
5/ solve


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com
 
Hello Drej

I ve just did it but temp is only localised in gypsum and steel temp is zero.
There is any conduction between those two materials.

What do you think
 
Have you checked the thermal properties you have specified for the steel? You may have forgotten to specify something by the sounds of it. You don't need to worry about conduction as long as at the boundary of the two materials the mesh is continuous.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com
 
For the thermal properties im editting
Thermal conductivity, Sp. Heat, Density.

For the mesh im using mesh areas->free->pick all

do you think that my mesh isnot continuous?

Thank you a lot Drej.
 
You'll have to check. Check the mesh at the boundary between the two materials - do the elements at the boundary share the same nodes? If not, then they need to.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com
 
maybe this is the problem,

could you tell me how those two elements can share the shame nodes please.

Please you will help me a lot

Thank you a lot Drej
 
Hi Juliet27,

I hope you have already solved your problem. As Drej said, it has probably to do with the elements at the boundary not sharing the nodes. You can visually check that by using the command:
/pnum,node,1
which plots the node numbers, followed by eplot. If the elements on the boundary don´t share nodes (that is, they have different numbers), then you can use the nummrg command in order to merge the nodes.

 
Status
Not open for further replies.
Back
Top