Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Some comments on pA material absolute pressure and CFD gauge & reference pressure

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

For those interested in pressure definitions

As I read often comments on the Forum, and I heard many questions on the last workshop at the Milan conference on CFD pressure, gauge, absolute, reference pressure and how and when to use them ...
I took some time today to see if I could find the description in the COMSOL doc and physics nodes entries. Yes it's all there (4.3a), somewhat scattered in all those pages, but I noticed that I got confused myself, but finally I believe I got some order on it, so why not comment it here too for other interested (and then I know where I can find it back too ;)

CFD and pressure:
This is so obvious for CFD experts that they do not think over it, but for most newcomers, deciding on which pressure to set in a CDF model is not that obvious. We are all used to work on earth at 1[atm] absolute pressure (about), but when we simulate some flow, often we are interested in cases with "p" of only a few [Pa] pressure drop, "p" compared to the "pA" 1[atm] or about 1E5 [Pa] that is a large difference. And, for those having studied numerical computations, making differences of large numbers that are similar in amplitude induces large relative error on the result. Therefore, CFD people mostly work with gauge pressures with p close to "0". The physics equations remain unconcerned if we add a constant value here, it solves as well, but with a small gauge pressure we are sure the results are more precise. Using an absolute outlet pressure of 1[atm] is certainly possible in COMSOL, but then do not forget o change also the initial value pressure to at least 1[atm] ... etc

Using a gauge pressure close to "0" [Pa] works well until we enter some material property that depends on the pressure, these will depend on the Absolute pressure pA, and we will get wrong results if we used simply "p" the gauge pressure.
Therefore: 1) when we define materials in the materials library, and these depend on the pressure, we should use the variable name pA (similar to T[1/K] that should be entered in, and arranged for, T in Kelvin and not °C, as K is COMSOL's default).
From the moment we call in a material property that is dependent on T or on pA, one will notice that COMSOL will populate, the "Model input" tab of the main physics sub-node, and here it will define pA as "pA = p_ref+p" (note this is for 4.3a I believe it has been adapted lately), where "p_ref" is a reference pressure, by default 1[atm]. In this way COMSOL calls for material properties with the correct absolute pressure and our results appear correctly ;)

You can find these explanations in the doc, partly in the CFD, partly in the materials doc. But also in the HT manual, as for thermal properties, as well as for ideal gas modelling (p*V=R*T), pA and T are required, in correct units and values, for all energy estimations

The gauge pressure, just as magnetic field gauge settings, voltage GND (gauge) settings ... are values one must often control, as else our PDE's are left in an unknown state with INF true and valid solutions

I hope these lines will help some of you too

And if I have missed something, or said something wrong pls update !

--
Have fun COMSOLing
Ivar

4 Replies Last Post 04.12.2012, 14:17 GMT-5
COMSOL Moderator

Hello Ivar KJELBERG

Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.

If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 03.12.2012, 10:16 GMT-5
Hi Ivar,

thank you very much for your explanations, they are always very helpful. It is right that if one is not yet an expert in CFD it is not clear at the first sight which pressure should be used.

At the moment I am working with laminar flow. So one more question: Did I get it right that if I define a reference pressure p_ref at the fluid properties another pressure p (initial values), p_entr (inlet) or p_0 (outlet) is just added to the p_ref? So for example if I need an absolute inlet pressure of 1 bar and the default pressure p_ref is already 1 bar the inlet pressure p_entr has to be set to 0 bar?
And what if I use the vacuum pump node? Is it the same case?

Thank you for your help.

Sandra

Hi Ivar, thank you very much for your explanations, they are always very helpful. It is right that if one is not yet an expert in CFD it is not clear at the first sight which pressure should be used. At the moment I am working with laminar flow. So one more question: Did I get it right that if I define a reference pressure p_ref at the fluid properties another pressure p (initial values), p_entr (inlet) or p_0 (outlet) is just added to the p_ref? So for example if I need an absolute inlet pressure of 1 bar and the default pressure p_ref is already 1 bar the inlet pressure p_entr has to be set to 0 bar? And what if I use the vacuum pump node? Is it the same case? Thank you for your help. Sandra

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 03.12.2012, 10:28 GMT-5
Hi

the pA is added to the local dependent "gauge" pressure "p" when referring to material properties that are normally now in the latest version using "pA" and not "p" for their internal variable dependencies. But the CFD is solving without the knowledge of pA since it's using the gauge pressure "p" close to "0".

In most cases on has a predefined outlet pressure p(x=L) = 0 which means in absolute pressure of pA+0 at output location "L", and then solve for an inlet flow velovity, hence COMSOL calculates the inlet pressure as a local pressure p(x=0) to be added again to pA to read the absolute value at inlet

--
Good luck
Ivar
Hi the pA is added to the local dependent "gauge" pressure "p" when referring to material properties that are normally now in the latest version using "pA" and not "p" for their internal variable dependencies. But the CFD is solving without the knowledge of pA since it's using the gauge pressure "p" close to "0". In most cases on has a predefined outlet pressure p(x=L) = 0 which means in absolute pressure of pA+0 at output location "L", and then solve for an inlet flow velovity, hence COMSOL calculates the inlet pressure as a local pressure p(x=0) to be added again to pA to read the absolute value at inlet -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 04.12.2012, 11:08 GMT-5
Hi,

in this case there should be a negative pressure p_exit used for the boundary condition vacuum pump?

Cheers
Sandra

Hi, in this case there should be a negative pressure p_exit used for the boundary condition vacuum pump? Cheers Sandra

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 04.12.2012, 14:17 GMT-5
Hi

I do not know of any limitations so long p_exit < pA ;)

but you can also define a p_in = 0 and a velocity flow at the output, no ?, but normally a pump has a v(p) or v(pA) dependency, all depends what your "pump" is

--
Good luck
Ivar
Hi I do not know of any limitations so long p_exit < pA ;) but you can also define a p_in = 0 and a velocity flow at the output, no ?, but normally a pump has a v(p) or v(pA) dependency, all depends what your "pump" is -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.