'Internal_Flow_Globe_Valve' simulation project by ashroff


#1

I created a new simulation project called 'Intrnal_Flow_Globe_Valve':

Flow through a globe valve


More of my public projects can be found here.


#2

Interesting geometry @ashroff!

Are you happy with the results or do you think you can improve the simulation or maybe even optimize the geometry?

Cheers!

Jousef


#3

Hi @jousefm, this was a preliminary simulation of my flow system. While the results do look promising, fine tuning of the model is indeed required. Post-processing with SimScale does seem to be a challenge though!

Cheers,
Aadit


#4

Hi @ashroff!

Absolutely no problem as we have some good resources as can be found here: Post-Processing Collection SimScale. Let me know if that was useful for you!

Best,

Jousef


#5

Hi Jousef,

Could you explain to me what would be the repercussions of changing the co-ordinates of the Material Point? My aim is to perform an internal flow simulation and although the location of my Material Point is changed from the default, it is still within the control volume. Would that affect my Mesh/Simulation in an adverse way? If so, how can I understand the theory behind the definition of the Material Point the way it is used in SimScale? Thank you!

Best,
Aadit


#6

Hi @ashroff!

The material point basically defines if you want to perform an internal or external flow simulation. If it is inside the model it will be an internal flow simulation and an external analysis otherwise assuming that the CAD model is watertight!

Does that answer your question?

Cheers!

Jousef


#7

Hi Jousef,

Thank you for your reply. That did clear it out for me more or less.
I had a certain doubt regarding the initial and boundary conditions for my steady state CFD simulation.

The Inlet pressure is 1200 psi (8273709 Pa) and inlet volume flow rate is 0.01m3/s. I have prescribed the volume flow rate and inlet velocity (computed from the inlet area, density and mass flow rate), the pressure at the outlet (0 bar gauge pressure) and of course no-slip walls as my B.C´s, while my initial conditions for pressure and velocity are both set to zero.

I am getting unsatisfactory results in the sense that the pressure generated in the body is far from the experimental values. I suspect that I have misunderstood the application of these conditions (I´m new to CFD, I work mainly with FEM). Could you perhaps take a look at my project and advise me on this?

Thank you!
Best,
Aadit


#8

Hi Aadit,

I am not at my machine at the moment and will have a closer look when I am home. @PowerUsers_CFD , any hints here?

Cheers!

Jousef


#9

Hi @ashroff,

Could you give a more detailed description of what you mean by “unsatisfactory results?”. Is the trend line similar? What about the results deviation? These will roughly allow us to determine where the likely source of error is.

Cheers.

Regards,
Barry


#10

Hi @Get_Barried,

In the light of recent results, I would like to refer you to my other project ''Choke Valve Flow Optimization". It is essentially the same project, but managed better.

My inlet pressure is fixed at the moment at 3.57 MPa (which is the safe operating pressure for the valve with water as the fluid), with inlet velocity and flow rate being 2 m/s and 16.2928 kg/s respectively. I do not know the outlet pressure nor the outlet velocity, but as there are no leaks in the domain, the mass flow rate at the outlet should be the same as the inlet. The goal of my simulation is to study the pressure drop across the valve body and further to optimize the geometry in order to reduce this drop.

In a few of my results, I find negative pressures at the exit of the valve body, whose physical significance I do not understand completely. The pressures computed shows a 10 fold increase in some regions of the valve body which should not be the case as it surpasses the safe operating conditions. With setting outlet pressure as 0 Pa (gauge pressure), I was hoping to view the difference in pressure between inlet and outlet directly from the pressure plots, but that is not the case as due to this condition the entire fluid volume is brought down to gauge pressure at the exit no matter what the inlet conditions are. I feel I am missing some vital piece of information at for the outlet, but I am confused so as to what they are. Let me know what you think of this. Thank you!

Cheers,
Aadit


#11

Hi @ashroff,

Apologies for the delayed reply, was a little busy over the past few days.

I’ve downloaded your data and post-processed it on ParaView so do take a look below.

As you can see, your flow is not behaving normally. There is essentially zero flow propagation within the valve itself due to the possible assignment of “freestream” as the inlet type. The fluid, in this case water, seems to be acting like a very dense form of air. What I would suggest is trying to set the inlet type to “Fixed Value” and see whether we can first get nominal flow behavior.

Cheers!

Regards,
Barry


#12

Hi @Get_Barried,

I took your advise into consideration, and have plotted the pressure and velocity fields in the same region as you did, but as you can see, the velocity magnitude is still zero throughout the range. I still am not sure about how to understand the negative pressures which i am getting.

I have used a custom exit boundary condition with the velocity gradient set to zero. Do you think specifying the exit flow rate should improve the results?

Cheers,
Aadit


#13

Hi Aadit,

Hmm strange. I’ll copy your project over and run it to see if I can get flow propagation. Will let you know if I get anything.

Cheers.

Regards,
Barry