[MGabr]

Dear all
now i am facing problem in reducing the values of Y plus , i am using SST turbulence model for simulating flow through an orifice in piping system. i am using first layer thickness 1*10^-5 and inflation rate 1.1 for maximum number of layers 20 layers , the results of Y Plus near 3000 which is absolutely wrong.

could you advice me please ?
if you need info. please let me know
best regards

[ghorrocks]

Why is y+ = 3000 wrong? What do you expect it to be? What are you modelling anyway?

Are you sure you are using the correct units?

[MGabr]

Dear ghorrocks
thanks for your reply , i think it is wrong because " as i know " it should be in rage of 1 for SST model , I am modelling flow at 6000 KPa passing through orifice in piping system. concerning your question about units , i am i didn't change ansys CFX default units.

best regards

[evcelica]

I think what Glenn is trying to say is how did you calculate y+ to know its not 3000?
A range of 1 with SST is just the recommended range, it doesn't mean SST will always give you a y+ of 1.

[MGabr]

evcelica; thank you for reply
your reply is very important , do you mean that if i found y+ values about 3000 , it
doesn't mean that my result is wrong. and if it doesn't mean wrong how can i define Y+ value should i stick to for my model ???

** note : I get Y+ values by getting the Y+ contour in the wall surface **
*** Just note : according to experimental data i have my error is about 9% in comparing mass flow rate results ***

best regards

[ghorrocks]

You need to do a sensitivity analysis to determine if your y+ is a problem. If your simulation does not have high sensitivity to the boundary layer then you are probably fine, but if it is sensitive you will probably need to go finer. But rather than guess what y+ you need do the sensitivity analysis and figure it out for your case.

[MGabr]

ghorrocks; I had did some kind of sensitivity analysis by changing the first layer thickness from .0004 to 10^-5 and i found almost no change " still in 3000 range " . do you suggest any other method to figure out its sensitivity.
If you have more than one method kindly mention it , I should proof that to my professor in my master degree thesis.

Thanks for your cooperation
Best regards.

[ghorrocks]

You have shown that a factor of 40 change in near wall resolution has almost no change in the results. That looks like pretty good evidence to me that you do not need to refine the near wall mesh further. For most applications I would be happy with that justification.

[MGabr]

ghorrocks ; thanks for your fast reply which is really appreciated. So according to our discussion and your opinion i would contact my professor and put this data between his hands . if he accepted that it would be nice ,and if not !! i would contact you asking for more suggestion

best regards
M.Gabr

[hellsblade.91]

Hi guys,


I was just reading this post cerning wall y value I am currently simulating air flow across my domain

The fluent documentation says while using k epilon model wall y+ value should be in the range of 30 to 300 to obtain accuracy in the simuln

I currently have a value of 3230 ( maximum) which certainly doesnt follow the criteria

Does that mean by results arent accurace I tried to change the mesh but then I get a floating error

[Gert-Jan]

M.Gabr,

how large is your orifice?
what is the material?
what is the pressure drop? Is it 6000 kPa?
or is the absolute pressure 6000 kPa?
what velocities do you obtain in your orifice?