Computational Fluid Dynamics (CFD) workbench using OpenFOAM

[Crossleyuk]

Yes, I did try the velocity input immediately after receiving that forum response, with no success.

However, making the change to the model case I've privided to you recently has resulted in success. Fantastic news!!

But does thie suggest that I can't repeat all the historic static pressure inputs that I successfully condicted with FC 0.17 and the old CfdOF? I'm trying to CFD existing physical testing that I conducted many years ago.

09-13-19 Images - Rev P1 Velocity Input.png (118.3 KiB) Viewed 2979 times

Many thanks and appologies if I've been difficult.

Michael

[oliveroxtoby]

But does thie suggest that I can't repeat all the historic static pressure inputs that I successfully condicted with FC 0.17 and the old CfdOF? I'm trying to CFD existing physical testing that I conducted many years ago.

Ah. I didn't realise that this setup was working previously. If you still have or can re-generate the OpenFOAM case that was created (the "case" directory that is written), that would probably tell us why.

[Crossleyuk]

Thanks Oliver, happy to do what ever I can.

I'm just now starting to migrate the 0.17 model case, that I've had great results from, into 0.18 and new CfdOF (both of which I love). It'll take me a day or two but my intention is to start from scratch and move forward. I have geat plans for this CFD assessment of my historic physicl testing data. With an aim of moving my research into a new era.

I'm loving the whole learning and relurning process and can't wait to try my revamped model. But will have to use static pressures for the two inputs and single output. My aim is to ontain CFD representations of the physical testing flow and non-dimensional pressure flow characteristics of the device.

Fingers crossed and I'll be back to you ASAP with as much information as I can.

Many thanks man!

Michael

[Crossleyuk]

Okay Oliver,

Rebuild complete with some changes to make it easy for me to check out the pressure/velocity imput implications.

The required model has four axial control inputs and four radial inputs. But I've now limited the radial inputs to one. The inputs are 1.0 m/s in the appropriate direction. There is still a single axial pressure output. I tried to included the model and free folder files. Going to look ito changing the output folder when I understand what is happening.

AdobeARM.log

(870 Bytes) Downloaded 71 times

VORTEX-CHAMBER-20190917-1105.log

(16.73 KiB) Downloaded 58 times

FC0.18 12in VSA Diffuser Only - Rev P1 Velocity Fail #1.PNG (88.64 KiB) Viewed 2903 times

Strange, I could not attatch the model file to this post. I assume that it's too big but no idea why. I'll take another look and get back to you.

Hope this is if some assistance and I'm still working on it and will get the model file to you. Would a pdf of the Report view text help?

Michael

[Crossleyuk]

After taking a closer look at the new dialogues for the fluid boundaries I noted that I could define a normal to wall and provide velocities at each of the four radial inlets in the attached model.

As a result, it now contains four radial and four axial inlets each with a 1m/s inlet velocity and it is working fine but is very slow. 500 iterations overnight.

Great progress though but I really must be able to define my actual testing approach, i.e. setting the following:

1) Atmospheric pressure at the four small axial inlets identified as "Axial_Control_Inlets"
2) Atmospheric pressure at the four larger radial inlets identified as "Radial_Inlet_#1 to #4"
3) A negative (fan) depression of -1500Pa gauge at the outlet identified as "Axial_Outlet"

Going to let it run to the 2000 steps to ensure all is good but will be making further changes,

Many thanks again.

OS: Windows 10
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.18.16117 (Git)
Build type: Release
Branch: releases/FreeCAD-0-18
Hash: dbb4cc6415bac848a294f03b80f65e888d531742
Python version: 3.6.6
Qt version: 5.6.2
Coin version: 4.0.0a
OCC version: 7.3.0
Locale: English/UnitedStates (en_US)

FC0.18 12in VSA Diffuser Only - Normal Velocity 500 Iterations.PNG (110.25 KiB) Viewed 2877 times

AdobeARM.log

(8.43 KiB) Downloaded 74 times

StructuredQuery.log

(18.3 KiB) Downloaded 72 times

aria-debug-8928.log

(470 Bytes) Downloaded 84 times

aria-debug-14116.log

(1.66 KiB) Downloaded 87 times

[Crossleyuk]

Here is the actual model for the above note. No mesh due to resulting large file size.

Michael

[Crossleyuk]

This was the final iteration screen after it failed with a memory error, probably my fault playing arround while it was running. Basically it worked great with velocity boudaries but will not run at all with the pressures boundaries I require.

I've also provided the text at the end of report view and some files from temp folder but not sure if they are valid to that case.

Michael

Report View - Momory Error.pdf

(8.52 KiB) Downloaded 70 times

aria-debug-10000.log

(470 Bytes) Downloaded 86 times

AdobeARM.log

(870 Bytes) Downloaded 71 times

[Crossleyuk]

Certainly working with velocity boundaries.

Glyph Image.PNG (564.71 KiB) Viewed 2820 times

Certainly working with velocity boundaries.

Michael

[Jee-Bee]

Cool

[Crossleyuk]

Hi Oliver,

OS: Windows 10
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.18.16117 (Git)
Build type: Release
Branch: releases/FreeCAD-0-18
Hash: dbb4cc6415bac848a294f03b80f65e888d531742
Python version: 3.6.6
Qt version: 5.6.2
Coin version: 4.0.0a
OCC version: 7.3.0
Locale: English/UnitedStates (en_US)

You were mentioned in dispatches because now I have a working FC 0.18 model I'm now having problems with Paraview and posted the following on Help on Using FreeCAD forum.

I have been using FreeCAD 0.17 and CfdOF for a year with great results. Recently moved to FreeCAD 0.8 and now having problems with Paraview 5.4.1. It will load from FreeCAD and from my Praview icon but I cannot do anything. It crashes and hangs up of I try to load a state or select open. It just sits there with no response and I have to end the task to shut it down. So I downloaded PV 5.4.1 but no change and then downloded PV 5.7.0 and it works fine. But I can't run it from CfdSolver or import my CfdOF case into this later Paraview version. So I'm confused, is there a way to get CfdSolver to start PV 5.7.0?

Response was:

You should probably ask in the FEM/CFD subforum. I know that the CfdOF project is under heavy development, so Oliver, the developer, generally recommends to try with the latest version of CfdOF, Paraview, and OpenFOAM, so that everything works okay. Also trying with the latest version of FreeCAD, 0.19.

So sorry, back to you again now.

First off, FreeCAD 0.19 is clearly not available but I see that 0.18.3 is. Basically, I can get Paraview 5.4.1 to work, hence the above image but I can't save or load states. I'm relluctant to make more changes but should I try 0.18.3?

Many thanks and apologies again.


Michael