[Woud]

Hi,

A couple of weeks ago i started with Openfoam, playing with the tutorials and watching Youtube videos. Now I am aiming to calculate the pressure drop over a section of pipe, which I will use later to calculate the pressure drop over some custom headers and bends (all PVC).

After the simulation using simpleFoam (starting from the squareBend tutorial) I ran the following two commands to determine the pressure loss:

postProcess -func "patchAverage(name=inlet,p)" -time 400

and

postProcess -func "patchAverage(name=outlet,p)" -time 400

resulting in 0.00331228 and 0.0 respectively. I know that I have to multiply this value with the density of the fluid (water) to get the actual pressure. This results in about 3.3 Pa of pressure drop over the pipe which is 2m long and 25mm in diameter and has an inlet flow of 0.1m/s.

Using http://www.pressure-drop.com/Online-Calculator/ i found that the pressure drop should be around 0.59Bar which is a massive difference.

I tried the refine the mesh but this didn't help. There are two directions I am not sure about, defining the roughness of the wall and the corresponding initial conditions and which solution schemes to use.

I attached the case, which hopefully helps with pushing me in the right direction.

btw, I am using Openfoam v2012

[DevilX]

Hey Woud,


i wans´t able to calculate your case myself, because my pc is running all time with my simulations.


My guess is, that your calculation on the website is wrong. I checked it with water and attached you find the screenshot with your geometry. Then the droop is 0,19 mbar, which sounds much better for me, because 0.59 bar pressure drop through a normal pipe ist really heavy.

[snak]

Hi,

pipe which is 2m long and 25mm in diameter

Is this correct? In your case, the diameter of pipe is about 200 mm. (It is estimated from a STL file.)


As Daniel pointed out, the pressure drop you obtained from the website will be incorrect. It will be the same order of your simulation.

[Woud]

Good evening,

Thank you Daniel for pointing out that I made an error in the pressure drop calculator, it should be 0.19mbar indeed.

@snak You are right as well, I tried to calculate the pressure drop simulation with a 200mm pipe first. (since this is closer to my final application) But after playing around with that and being about 30% off the value I expected I tried a smaller diameter. Now it seems like I did not delete the old .stl file properly, running my simulation with the 200mm pipe.

I corrected the mistakes and now I have a pressure drop of 47mbar, which is already much better and in the correct range of magnitude. However, it is still more than twice the value expected.
Should I learn to life with that or can I improve with choosing another solver or tweaking some of the conditions?

[DevilX]

Well try to reduce the limitation factor in your schemes, maybe then you accuracy rises.
Twice the difference is still huge. Try the schemes and also make your Mesh finer (Just a try - the computational time should not be a problem here). Furthermore, you can try differnent Turbulence schemes as well. Also try to verify your pressure drop theoretically for yourself (pipe roughness f.e.).

[Woud]

Oke, I found some time to work with your suggestions.
Starting with the easiest thing first, refining the mesh. Which didn't help at all.

During one of my google searches I noticed that i think that my problem is a steady state problem, however the turbulent behaviour can introduce vortexes which makes the problem transient? Therefore, simpleFoam is not the right solver.

Furthermore,
- I verified the pressure drop with some basic calculations resulting in similar pressure drops as from the online calculator. Since I am in the transition region between laminair and turbulent the friction factor gives a significant difference (laminar dP = 22.4Pa and turbulent dP = 44Pa).
- I changed the length of the pipe from 2 to 5 meters, to have a bit more significant pressure drop to work with.

I have been playing with some of the schemes and limitation factors, but I am not sure what I am doing.
Therefore, I added my case again. Hopefully some of you can push me a bit more in the right direction