slice and avg air velocity. the inlet result is different from inlet settings

Dear all,

I’m new to paraview and I’m spending a lot of positive energies discovering the cfd and post processing work.

For learning process, I’m investigate a ventury effect on a pipe and I want to see difference on flows between inlet and outlet. Design and computation arranged by Freecad and OpenFOAM

I read few posts around about resampling and so on. for the information I’m looking for it seams slice will do great because I’m only looking for the general avg U

Input specs: fluid, air; radius,60mm; flow, 25l/s. calculate avg air velocity: 2,12m/s

I slice@ 0.001m from the inlet surface then apply filter integrateVariables and the results are confusing.

in the spreedshet, if I visualize cell data I found and area of 0,0113m2. when I chech the U either cell data or point data, the magnitude i read is of 0,0248 (should it ring any bell that is close to the air flow of 25l/s?)

in the graphical view, if I plot U_mag and adjust U velocity scale, the scale show mas velocity of 2.2 m/s meaning that there is some ‘confusion’ in the integration or at least in my understanging of it.

Botttom line, if it is ‘correct’ that the U_Mag shows the flow rate, I can find the avg velocity but then the question is why this number still appear under U field.

I use paraview 5.10.0-RC1

any help or hint is much appreciated

thanks

You may want to update to the last release of ParaView

Please share some data and screenshots of the issues you are seeing

Hi @mwestphal , thanks for your prompt reply.
Regarding PV version, this is the default innstalled in ubuntu using Synaptic. I’ll purge and install the lastes from source. thanks for the advice.

Below the screenshot of the inlet slice. The spreadsheets refers to the inlet.

The slice is taken 1mm inside the face.

inlet radius is 60mm with an area of 0,0113m2 if we make a simple calculation.
The inlet speed, based on calculation, will be 0.025(m3/s) / 0.0113 equal to 2.212m/s which match the U scale but not the value in spreadsheet

The possibility that the spreadsheet is calculating the flow it is pure speculation based on number but definetively is not the velocity

I hope this description makes situation clearer.

thanks

[extra]
i just changed dimention setting inlet flow of 15l/s. Also in this case, using same procedure, the u_magnitude is 0,0144 (close to the inlet flow with some model approximation

Just use the binary release from https://paraview.org/download

I hope this description makes situation clearer.

Not at all, please explain what result is wrong in ParaView and why do you think it is wrong.

ciao @mwestphal

the theretical inlev velocity, based on flow rate and inlet surface, is approx 2,2m/s which is consistent with the U_Magnitude scale in the attached picture.

On the other hand, the U-Magnitude shown in the spreadsheet from the same attached pic is showing 0,02499 which is not accurate if we consider it being speed but accurate if we see it for some reason representing the flow rate. if it is the flow rate, the average speed can be easily calculate it is just matter of label of the spreadsheet

Cheers

All integrate variable do is suming the values, then dividing by the surface, so this computation may not be what you expect in that specific case.

I think we should move this conversation to a different level possibly bug or feedback, please sugest

First of all, the result listed as U_magnitude which is a speed, is wrong because we know the theoretic value confirmed by the graphical scale and this is clear

second, accordingly to your comment, the filter should sum the velocities then devide by the surface which is again something not correct in this situation. The concept of integrating the flow for each cell involved by the slice then devide by the surface make sense. the actual situation in this specific case it doesn’t devide the sum of flows by the slice area showing, as supposed, the actual total inlet flow.

As a newbi and generally speaking, because paraview doesn’t shows dimension, we have to trust that what is listed being a speed it is really a speed and not something else. This is why I suggest to report this discussion to whoever manage the integration filter.

I hope it is clear my intention is to bringing constructive feedback and not want to sound bad under any circumstance.

Available for further discussion

thanks

Here is where to report issues: https://gitlab.kitware.com/paraview/paraview/-/issues

Try activating the “Divide Cell Data by Volume” tick box option in the integrate variable filter.

Alexandre

Ciao @Alexandre_Minot ,

it seams your trick works or to be more precise the number shown after ticking the “Divide the cell data by volume” match the one I found if I devide the U_magnitude without ticking and divide by cell area

Definetively this way of showing results is misleading because if the filter shows a number in the velocity field it should be compatible with phisical dimention. there is a great difference between velocity and flow.

Again, my comment are just to warn those who are stacked in similar situation and to avoid confusion as it happens to me.

Thanks everybody for your attention. Great job!