CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Perforated plate (https://www.cfd-online.com/Forums/cfx/88515-perforated-plate.html)

 d_A_n_G May 19, 2011 05:27

Perforated plate

Hello, I have problems in modelling this geometry in ansys. How can I define the conditions? Is there any tutorial??

http://img34.imageshack.us/img34/421...ratedplate.png

This is like 120 holes with 20mm diameter but 48 times per product!

 ghorrocks May 19, 2011 07:40

What are you modelling, what are you trying to get out of the model and what are the physics you are modelling?

 d_A_n_G May 19, 2011 08:15

I'm modelling the hot air flow inside of an industrial drying process. Here's the problem in question:

http://img8.imageshack.us/img8/4213/problem4f.png

I simplified the 120 holes by 6 holes (per intake) with same total area.

I suppose that this assumption might not be correct in terms of air inlet difusion (is better uniform with 120 holes). Also for the speed, which might large influence the flow.

Once each inlet has different thermo conditions, I simulated both the two inlets and the two outlets separately, and then import into a global simulation:

http://img37.imageshack.us/img37/7098/real1setup.png

Then I obtained:

http://img59.imageshack.us/img59/4544/real1temp1.png

Objective is:
To uniformize the air inlet in such a way that the temperatures inside are uniform (we can see the speed differences in the inlet).

Hope it helps...

My next step is to attribute to the walls, the temperatures (once it creates the hot internal atmosphere...)

Thank you.

 ghorrocks May 19, 2011 08:21

So isn't the problem just that the inlet ducting does not distribute the hot air evenly?

 d_A_n_G May 19, 2011 08:26

In the current situation, yes, the hot air is not well distributed. But once also the extraction is not uniformized through all the "hoven", I have to play with this differences.

I think I cannot simply uniformize the inlet speed (by having the same drop pressure in all ducts compared to the main duct). And why? Both at the begining and end of the "hoven" there is a 'dead' zone without neither intake and neither outake. In these areas I might need more inlet speed and at the middle of the "hoven", less inlet speed once the atmosphere is already hot, and compensates the differences.

Objective is to play simultaneously with the inlet speeds and outlet speeds to uniformize the temperature profile. Is clear?

 ehsan_am86 May 19, 2011 10:31

share skill

Quote:
 Originally Posted by d_A_n_G (Post 308318) Hello, I have problems in modelling this geometry in ansys. How can I define the conditions? Is there any tutorial?? http://img34.imageshack.us/img34/421...ratedplate.png This is like 120 holes with 20mm diameter but 48 times per product!
Hi,my master thesis is about simulation of perforated tray and i am simulating this with in CFx.
i think that we can share our skills between us in order to improvement our work.could i have your email?

 Adam S May 19, 2011 11:34

You can´t substitute the holes by a permeability face?

 d_A_n_G May 19, 2011 11:48

I think so, but I never found how to assign that condition in a surface..

 ghorrocks May 19, 2011 18:58

I think you need to clarify exactly what you are trying to do. It seems to me you have two separate projects here.

It seems the inlet duct needs improvement to make the flow more even. Simulations can be done to achieve this simply modelling the inlet ducting. You can probably model the small holes as a resistance to get the back pressure right.

Then a separate model is to see how even the conditions in the dryer are. You could use the flow rates and velocities modelled from the previous model as inlet boundaries for this model.

This will be much more easily dealt with if modelled as two separate systems. Only model it as a single full system if they are coupled somehow.

 d_A_n_G May 23, 2011 06:14

Hello, thank you for your support and sorry for late back.

Yes indeed, I have two separate problems. In one hand, I have to make outlet more even (in the holes), and on other hand I have to uniformize the longitudinal profile temperature.

I'm currently doing separate simulations for the inlet duct, by closing the inlet area for each group of holes... I already found a kind of closing % for each component. I already changed it in the real case (in the placement), and I already got two uniform temperatures, out of three (2/3).

Before closing:
http://img35.imageshack.us/img35/812...velstreaml.png

After closing:
http://img199.imageshack.us/img199/4...streamline.png

However, the second inlet is higher temperature than the first, and once the extraction is not uniformized (more in the centre), I have to play with all variables to uniformize temperature.
Look at the different inlet and outlet speeds:
http://img855.imageshack.us/img855/8...lltempvel2.png

Once the extraction is more concentrated in the middle areas, there is less concentration of hot temperatures. it means that in the sides, i might need less inlet speeds once also there is less outlet speeds...

One question: how can I model the small holes as a resistance pressure to get back??

You said:"You could use the flow rates and velocities modelled from the previous model as inlet boundaries for this model" ---->>>>> that's what I did! i simulated the four ducts separately (the two inlet and the two outlet) and then imported to one single global simulation.

My big problem: Despite all the flow simulated behing, there is one important thing: the air infiltrated.
In the real case, the inlet mass is lower than the outlet mass rate (depression condition in the dryer). Then, there is air get infiltrated into the dryer (cold air 25ēC)... And the results give me that the infiltration speeds are higher and are influencing the temperatures.
This picture shows the temperatures with the infiltration (in the left side):

http://img804.imageshack.us/img804/9...temptemp11.png

A cold mass air entering in the left side and reducing the second zone temperature...

 All times are GMT -4. The time now is 10:50.