water flow simulation
Moderator: oliveroxtoby
Forum rules
and Helpful information for the FEM forum
and Helpful information for the FEM forum
water flow simulation
Hi Guys,
I simulate the nozzle from an abrasive water cutting system. For me, the speed distribution and the pressure distribution is important. There is a pressure of 1000 bar at the inlet (28 m/s velocity, 2.3 * 10 ^ -4 m^3/s volumetric flow rate). Inlet 1 is only one hole at which ambient pressure is present (defined as "open"). The outlet is the small diameter at the end (outlet). When I compare my simulation to reality, the results don't match. Theoretically, the water jet has to "shoot" through the nozzle because the pressure and speed are high and a negative pressure forms in the mixing chamber. My jet always swirls in the mixing chamber, which shouldn't be. Can anyone help me further? I tried laminar and turbulent flow, the result was the same for both.
for example:
inlet: 28 m/s
inlet1: open (ambient pressure)
outlet: static prssure
in the picture: red arrow (reality) , black arrow (simulation)
File with mesh:
https://www.dropbox.com/s/apl97dwndqxv9 ... FCStd?dl=0
Thank you
I simulate the nozzle from an abrasive water cutting system. For me, the speed distribution and the pressure distribution is important. There is a pressure of 1000 bar at the inlet (28 m/s velocity, 2.3 * 10 ^ -4 m^3/s volumetric flow rate). Inlet 1 is only one hole at which ambient pressure is present (defined as "open"). The outlet is the small diameter at the end (outlet). When I compare my simulation to reality, the results don't match. Theoretically, the water jet has to "shoot" through the nozzle because the pressure and speed are high and a negative pressure forms in the mixing chamber. My jet always swirls in the mixing chamber, which shouldn't be. Can anyone help me further? I tried laminar and turbulent flow, the result was the same for both.
for example:
inlet: 28 m/s
inlet1: open (ambient pressure)
outlet: static prssure
in the picture: red arrow (reality) , black arrow (simulation)
File with mesh:
https://www.dropbox.com/s/apl97dwndqxv9 ... FCStd?dl=0
Thank you
- Attachments
-
- 2D_Test_Velocity_without_mesh.FCStd
- (43.68 KiB) Downloaded 76 times
-
- 2D_Test_U_1.jpg (71.65 KiB) Viewed 3022 times
Re: water flow simulation
You could try closing the open first to see what it looks like.
-
- Veteran
- Posts: 3129
- Joined: Sat May 20, 2017 12:06 pm
- Location: Germany
Re: water flow simulation
Can you set the smallest cross-section from Pos1 to Pos2?
Smallest cross-section at Pos1 induces turbulence.
As Johnwang said: close the other inlet for a control-run.
Smallest cross-section at Pos1 induces turbulence.
As Johnwang said: close the other inlet for a control-run.
Re: water flow simulation
same settings and inlet1 closed. still doesn't look good...
- Attachments
-
- 2D_Test_Velocity_turbulent.png (49.22 KiB) Viewed 2963 times
-
- Veteran
- Posts: 3129
- Joined: Sat May 20, 2017 12:06 pm
- Location: Germany
Re: water flow simulation
Theoretically...
I played with the parameters:
Closed Inlet 2. Changing viscosity, inlet speed, mesh-refinement.
But no chance, the jet does not "shoot" through the chamber.
Is this an observed effect or a speculation?
Re: water flow simulation
Theoretically, reduce area to increase velocity.
There are 4 size of pipes from left to right before the nozzle. You can shoot from Size 1 pipe into Size 2 pipe. From Size 2 pipe to size 3 pipe, you lost a bit velocity. Form 3 to 4, lost more. Though pressure is build up in pipe 4 (the mix chamber), but pipe 1 is just too small to power it up to shoot through the nozzle.
There are 4 size of pipes from left to right before the nozzle. You can shoot from Size 1 pipe into Size 2 pipe. From Size 2 pipe to size 3 pipe, you lost a bit velocity. Form 3 to 4, lost more. Though pressure is build up in pipe 4 (the mix chamber), but pipe 1 is just too small to power it up to shoot through the nozzle.
- adrianinsaval
- Veteran
- Posts: 5541
- Joined: Thu Apr 05, 2018 5:15 pm
Re: water flow simulation
Is it even reasonable to expect acurate results from a single phase simulation of a two phase real life situation? was this not discussed in another topic were multiphase was already suggested?
Re: water flow simulation
@thschrader
it is a speculation, but in reality it is the case that the jet shoots through the mixing chamber and creates a negative pressure there.
@johnwang
do you think it will work if i increase the smallest diameter?
@adrianinsaval
We have already discussed the two-phase topic in this post.
https://forum.freecadweb.org/viewtopic. ... 5&start=20
to simplify it:
My model is just a tube with a side hole. Water flows through this pipe at high speed. So I need for this case the interfoam solver?
it is a speculation, but in reality it is the case that the jet shoots through the mixing chamber and creates a negative pressure there.
@johnwang
do you think it will work if i increase the smallest diameter?
@adrianinsaval
We have already discussed the two-phase topic in this post.
https://forum.freecadweb.org/viewtopic. ... 5&start=20
to simplify it:
My model is just a tube with a side hole. Water flows through this pipe at high speed. So I need for this case the interfoam solver?
-
- Posts: 991
- Joined: Wed Mar 27, 2019 10:45 am
Re: water flow simulation
You need two phases. That's a definitive.
For those still doubting the water jet creates air entrainment. By taking the air with it through the focusing tube, it depletes the air in the intake chamber. Therefore it creates a depression, sucking air from the air inlet.
For those still doubting the water jet creates air entrainment. By taking the air with it through the focusing tube, it depletes the air in the intake chamber. Therefore it creates a depression, sucking air from the air inlet.
Re: water flow simulation
The simulation does not reflect reality, because under such conditions there will be cavitation. This type of simulation will create pressures below absolute zero in the nozzle, which can not happen in reality. Just have a look at the pressure. So there should also be water vapor in some parts of the mixing chamber. Openfoam has a solver for cavitating flow. You could give it a try, but you need to set up a case directly in openfoam. Another question is the influence of gravity and air intake as others already said.
Ulrich
Ulrich