[giangcoikx]

Hi everyone,
I am a newbie in this area of using FLUENT. At the moment I am struggling with my numerical experiment such that I do not know my source of error, I am very happy to get you guys's help.

I am trying to measure our airship's drag. Here are its geometry characteristics:
Length L = 6.5m,
Max diameter D = 1.75m
Just bare hull, without tails.
Velocity =15m in y direction

I have tried with both scaled model and full-size but I have got the same bad results . It is ridiculously low, near zero meanwhile on book;s reference, its volumetric Cd is typically from 0.02 to 0.03. So I post here every my set-up steps of full-size experiment for simplicity (because for scaled model, it is necessary to adjust density, pressure in order to maintain the same Re and M numbers of dynamical similarity). Could you help me find out sources of errors?

About my computational domain:
In front and aft the body: 30m (approximately 5L), blockage ratio = 1%
Here I also attach the picture of pressure distribution on the pressure outlet's plane. I think it agrees well with my expectation, there is only small portion of negative gauge pressure. Am I right that it means a small backflow? Does it affect considerably to Cd? Should I extend it more?
(I set Cd as volumetric Cd, proportional to Volume^2/3, typically for airship.)
Operating environment are in sea level standard condition. So operating condition is 101325 Pascal.
Both estimation and calculated y plus are less than 1. (I have chosen k-w SST and Spalart model)
I have created the wake mesh for wake resolution as well.
The outer wall, I set it as zero-slip wall.

There are a few cells of bad equivolume skewness on outer vicinity of the wake region, anything else are fine. I guess that it does not affect to the hull's boundary layer calculation, because on the scaled model mesh, everything regarding mesh's quality is good but the results I got are still terrible.
Of course, I have calculated until all equations converged, most of them below 10^=6, the rest are about 10^-4, 10^-5, but the Cd converged to the wrong place, about 0.01.
All under relaxation factor are applied.
Solution initialization is set from inlet condition.
About the warning in the user-text, I do not understand thoroughly what it means. could you explain to me. I also have checked "cell wall distance" and attach it here.

I have tried my best to review everything including theory, but I have not found any source of errors yet. So I decide to ask experts.

Thank you so much.

[giangcoikx]

I continue attaching my setup's pictures here.

[giangcoikx]

I continue attaching my setup's pictures here. Also there is a force report, it is impossible for such a low drag with 6m long airship.

[giangcoikx]

Let me show the mesh in more detail:
My hull is divided into 2 symmetric region, I mesh half of the hull. The vertical symmetry plane is Oxy plane, the horizontal symmetry plane is Oyz plane. The velocity direction is in +y direction.
It means force in x and z direction should cancel each others due to symmetric reason. The +y direction force is drag. However, as the file txt shows:
the x-force is about 0.066, approximately 0, but the z force is about 157.54 N, I think it is meaningless.
The y force is less than 9, in terms of volumetric drag coefficient, its Cd = 0.011. Meanwhile for others airships, their volumetric Cd are about from 0.02 to 0.03.

In addition, the low viscous force is equivalent to low surface area, lower than our airship's surface area.
Could you guys help me explain such a weird phenomena?

Another question is: when I apply "symmetry" type for symmetry surface, so the drag reported here is for half-hull or for full hull? So does the drag coefficient? Because I enter volume^2/3 as reference area. If the drag is calculated for half-hull, my drag coefficient is meaningful, otherwise, it is meaningless