Total pressure loss

Shljuki · June 10, 2013, 05:35

Hi, i am simulating flow through a duct, two ends of the duct (inlets) are open to external space (ambient condition with relative pressure of 0 pa) and third end is outlet with 2.2 kg/s of air. I am interested in velocity distribution inside the duct and total pressure loss.
I am bit confused with total pressure distribution isnide the duct, it drops from starting 0 to -900 and starts increasing once the flow approach outlet boundary. I expect constant drop (loss) of total pressure. I have run simulation with different options for outlet: constant flux, shift presure, scale mass flows.
Can someoene advise why do i have total pressure increase at the end of the duct, close to boundary outlet?

oj.bulmer · June 10, 2013, 06:56

Can you post an image

OJ

Shljuki · June 10, 2013, 08:49

I dont have my CFD files here, attached is a schematic of the problem - hope it helps, if not i will be able to upload better image tomorrow.
Thanks

singer1812 · June 10, 2013, 12:34

Are you measuring Total Pressure along a stream line?

oj.bulmer · June 10, 2013, 12:44

Are you sure the solution is converged? Also, are you seeing area averaged values?

OJ

Shljuki · June 10, 2013, 20:05

The solution seem to be converged to below 10e-4, i am monitoring 4 different parameters and no changes.

I have been checking area average total pressure at different sections of the duct.

singer1812 · June 11, 2013, 10:05

Use massflow average. Or measure the total pressure at different points along a streamline.

oj.bulmer · June 11, 2013, 11:07

Contrary to what NS equations predict, it has been documented that in viscous flows, there is a possibility that total pressure may rise locally and this is a result of the redistribution of energy by the viscous stresses.

http://adsabs.harvard.edu/abs/1995AIAAJ..33..772I

The phenomenon is aggravated if models like k-eps are used since they sometimes over-predict the viscosity further aggravating this problem. And thus the modelling assumptions make this situation worse and this is the likelier reason for inconsistencies.

What model/discretization schemes are you using? Is the solution mesh independent?

OJ

Shljuki · June 12, 2013, 23:45

Thanks everyone for your support, mass flow average helped understanding results.
Total pressure increase along a streamline has ben recorded at the location where two air streams at about 25 m/s clash and enter third vertical duct - refer to attached images.

oj.bulmer · June 13, 2013, 08:41

Exactly, it has been documented that the rise in total pressure can happen due to reasons mentioned in above post, at stagnation point:
http://fluidsengineering.asmedigital...icleid=1429388

Now, in your case, where the two fluids meet and suddenly come to halt, it seems to be equivalent to what happens at a stagnation point and this may explain the increase.

However, you need to do a bit of literature search to understand if the extent of increase you see there is adequate or too much!

OJ

June 10, 2013, 05:35	Total pressure loss	#1
Shljuki New Member Join Date: Jun 2012 Location: Melbourne Posts: 14 Rep Power: 14	Hi, i am simulating flow through a duct, two ends of the duct (inlets) are open to external space (ambient condition with relative pressure of 0 pa) and third end is outlet with 2.2 kg/s of air. I am interested in velocity distribution inside the duct and total pressure loss. I am bit confused with total pressure distribution isnide the duct, it drops from starting 0 to -900 and starts increasing once the flow approach outlet boundary. I expect constant drop (loss) of total pressure. I have run simulation with different options for outlet: constant flux, shift presure, scale mass flows. Can someoene advise why do i have total pressure increase at the end of the duct, close to boundary outlet?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Error in run Batch file	saba1366	CFX	4	February 10, 2013 02:15
Total pressure in CFX	famarcfd	CFX	0	June 17, 2011 11:33
definitioon of total pressure with k	Carl H.	Main CFD Forum	4	June 7, 2005 04:59
Total Pressure in Laval Nozzle	Sohail Ahmed	Main CFD Forum	0	May 19, 2004 06:43
what the result is negatif pressure at inlet	chong chee nan	FLUENT	0	December 29, 2001 06:13

June 10, 2013, 06:56		#2
oj.bulmer Senior Member OJ Join Date: Apr 2012 Location: United Kindom Posts: 473 Rep Power: 20	Can you post an image OJ

June 10, 2013, 12:34		#4
singer1812 Senior Member Edmund Singer P.E. Join Date: Aug 2010 Location: Minneapolis, MN Posts: 511 Rep Power: 21	Are you measuring Total Pressure along a stream line?

June 10, 2013, 12:44		#5
oj.bulmer Senior Member OJ Join Date: Apr 2012 Location: United Kindom Posts: 473 Rep Power: 20	Are you sure the solution is converged? Also, are you seeing area averaged values? OJ

June 10, 2013, 20:05		#6
Shljuki New Member Join Date: Jun 2012 Location: Melbourne Posts: 14 Rep Power: 14	The solution seem to be converged to below 10e-4, i am monitoring 4 different parameters and no changes. I have been checking area average total pressure at different sections of the duct.

June 11, 2013, 10:05		#7
singer1812 Senior Member Edmund Singer P.E. Join Date: Aug 2010 Location: Minneapolis, MN Posts: 511 Rep Power: 21	Use massflow average. Or measure the total pressure at different points along a streamline.

June 11, 2013, 11:07		#8
oj.bulmer Senior Member OJ Join Date: Apr 2012 Location: United Kindom Posts: 473 Rep Power: 20	Contrary to what NS equations predict, it has been documented that in viscous flows, there is a possibility that total pressure may rise locally and this is a result of the redistribution of energy by the viscous stresses. http://adsabs.harvard.edu/abs/1995AIAAJ..33..772I The phenomenon is aggravated if models like k-eps are used since they sometimes over-predict the viscosity further aggravating this problem. And thus the modelling assumptions make this situation worse and this is the likelier reason for inconsistencies. What model/discretization schemes are you using? Is the solution mesh independent? OJ

June 13, 2013, 08:41		#10
oj.bulmer Senior Member OJ Join Date: Apr 2012 Location: United Kindom Posts: 473 Rep Power: 20	Exactly, it has been documented that the rise in total pressure can happen due to reasons mentioned in above post, at stagnation point: http://fluidsengineering.asmedigital...icleid=1429388 Now, in your case, where the two fluids meet and suddenly come to halt, it seems to be equivalent to what happens at a stagnation point and this may explain the increase. However, you need to do a bit of literature search to understand if the extent of increase you see there is adequate or too much! OJ