[evcelica]

Greetings,

I am modeling a liquid domain which has a perforated plate. I know the streamwise directional loss coefficient but am having trouble estimating a realistic transverse loss coefficient. right now I'm just using 1/10 the streamwise coefficient, but am trying to see if I cane comehow describe it better.

Has anyone seen any good resources/papers to calculate the transverse loss coefficient, or have any experience?

I found one: "Compact CFD Modeling of EMC Screen for Radio Base Stations: A Porous Media Approach and a Correlation for the Directional Loss Coefficients", but it is for air and the study is very different than my flow characteristics.

Thanks in advance

[ghorrocks]

A perforated plate sounds thin. In that case there would be little thickness for the transverse flow to do anything. This means the transverse coefficient will not matter much.

I recommend you do a sensitivity analysis of it, and I suspect you will find that it will not make much difference, so you can enter anything within a wide range of the streamwise loss coefficient.

[evcelica]

Thanks for the reply,
That was my thinking that it shouldn't have much effect on the transverse flow. I'll do the sensitivity study with 0%, 10% and 20% the flow restriction of the streamline direction.

[zegtuhetmaar]

Did you have any result? I agree with flows coming with no angle of attack, but what about flows coming in with an angle of attack. Then it makes a difference when choosing directional loss.

My question is what should be this factor for 45º flows?

[oj.bulmer]

Originally Posted by evcelica

I am modeling a liquid domain which has a perforated plate. I know the streamwise directional loss coefficient but am having trouble estimating a realistic transverse loss coefficient. right now I'm just using 1/10 the streamwise coefficient, but am trying to see if I cane comehow describe it better.

What is your t/d ratio? t being the thickness and d being the hole dia. If it is less than 0.015, you needn't worry about the transverse loss coefficient and just model it using the porous interface.

If it is more than 0.015, it is regarded as thick plate. And generally the transverse loss coefficient will be far larger than streamwise coefficient for thicker plates. I don't understand why you say that you will use 1/10th of the streamwise...

Originally Posted by zegtuhetmaar

My question is what should be this factor for 45º flows?

How angle affects the loss coefficient strongly depends on t/d ratio, with perfs having larger t/d ratios affected more than smaller ones. Typically, upto 30 deg, the loss coefficients don't seem to affect much, but beyond that, it increases rapidly, the rate depending on t/d ratio, of course.

OJ