CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   buoyancy - losses estimation through total pressure (https://www.cfd-online.com/Forums/cfx/81545-buoyancy-losses-estimation-through-total-pressure.html)

 clrsgfn October 29, 2010 14:07

buoyancy - losses estimation through total pressure

Hi everyone

As a test, I'm simulating a basic vertical cylinder, with uniform normal water flow at entrance and zero averaged pressure at outlet. Reynolds number is ~2000

- When buyoancy is off, by calculating total pressure difference between inlet and outlet I get losses that corresponds to what can be expected from theory

- when buoyancy is on : computing losses by total pressure difference gives a very different value (*20 !). Computing losses by difference of (abs pressure+cinetic energy) gives non physical results (negative losses...)

I've checked hydrostatic pressure (abs. pressure - ref. pressure - pressure) and this is ok...

what's going on ? and how do I compute losses in case of buoyancy ?

thank you all in advance !

 TommySean October 30, 2010 11:32

If there's setting about reference density, you can make a test with reference density = 0.

 clrsgfn November 1, 2010 14:03

thanks
but I am not considering a multiphase flow, as only water is flowing in. So my density is fixed and constant (i am not considering temperature fluctuation either). This is really a simple test case, and results are puzzling to me...
any other clue ?

 michael_owen November 4, 2010 13:10

Quote:
 Originally Posted by clrsgfn (Post 281420) Hi everyone As a test, I'm simulating a basic vertical cylinder, with uniform normal water flow at entrance and zero averaged pressure at outlet. Reynolds number is ~2000 - When buyoancy is off, by calculating total pressure difference between inlet and outlet I get losses that corresponds to what can be expected from theory - when buoyancy is on : computing losses by total pressure difference gives a very different value (*20 !). Computing losses by difference of (abs pressure+cinetic energy) gives non physical results (negative losses...) I've checked hydrostatic pressure (abs. pressure - ref. pressure - pressure) and this is ok... what's going on ? and how do I compute losses in case of buoyancy ? thank you all in advance !
The absolute pressure is equal to the reference pressure plus the pressure. P_abs - P_ref - P is not P_hydrostatic, it's zero. The reason your pressure difference is so large is that the Pressure includes the hydrostatic pressure. What is the height of the cylinder? You have a hydrostatic pressure difference of rho*g*h. Subtract that from delta P and you will get your losses.

 clrsgfn November 10, 2010 12:05

Quote:
 Originally Posted by michael_owen (Post 282176) The absolute pressure is equal to the reference pressure plus the pressure. P_abs - P_ref - P is not P_hydrostatic, it's zero. The reason your pressure difference is so large is that the Pressure includes the hydrostatic pressure. What is the height of the cylinder? You have a hydrostatic pressure difference of rho*g*h. Subtract that from delta P and you will get your losses.
Hi michael,
thanks for your contribution.
I've understood that when buoyancy is on, absolute pressure = pref+pressure+hydrostatic pressure (this is what is said in CFX help).
In fact I did a wrong calculation when buoyancy was on as my reference temperature was set in kelvin and not in celsius as I thought. Now that I fixed it, total pressure difference when buoyancy is on or off are the same.

Therefore, I still have questions about hydrostatic contribution :
when buoyancy is on, total pressure difference as defined by CFX (pressure without hydrostatic contribution + cinetic energy) between inlet and outlet should be regular losses + rho.g.deltaz
and this is not the case, we get only regular losses.

Could it be that when imposing boundary conditions, hydrostatic contribution is implied, therefore in the pressure field obtained, hydrostatic contribution is indeed included in PRESSURE ?

Thanks.

 All times are GMT -4. The time now is 04:36.