-
Filter by Topic
Most Popular
All Topics
- List all discussions
Facebook
Twitter
Delicious
Digg
Stumbleupon
More services...sinosoidal inlet boundary condition
|
Thread index | Previous thread | Next thread | Start a new discussion |
RSS feed |
Turn on email notifications |
5 Replies
Last post: July 31, 2010 5:27pm UTC
July 29, 2010 9:37pm UTC
sinosoidal inlet boundary condition
hi
I want to emply the inlet velocity as boundary condition in the form of sinusoidal function(A+Bsin(wt)).I wrote the expression in the options>Expressions>Scalar expression.but when it begins to run I get the error massage as:
"not able to find inconsistent initial value"
but when I change the function as a polynomial it is solved.
Dose anyone know what is the problem in entering the inlet boundary condition as sinusoidal function?
whould u please help me..
I would be very thankful
maral
Reply | Reply with Quote | Send private message | Report Abuse
July 29, 2010 10:49pm UTC in response to maral hassani
Re: sinosoidal inlet boundary condition
hi
I want to emply the inlet velocity as boundary condition in the form of sinusoidal function(A+Bsin(wt)).I wrote the expression in the options>Expressions>Scalar expression.but when it begins to run I get the error massage as:
"not able to find inconsistent initial value"
but when I change the function as a polynomial it is solved.
Dose anyone know what is the problem in entering the inlet boundary condition as sinusoidal function?
whould u please help me..
I would be very thankful
maral
I can't imagine that the problem is the sin function, but the numerical value of it. Did you plot your inlet function (in alpha wolfram e.g.) to see if it starts from 0? Otherwise the numerical solver will have problem converging the first steps.
If your flow is always at high velocity, then you may take another approach. First ramp your inlet velocity and then use your normal oscillating velocity:
inlet = flc2hs(t-.05,.1)*Umax*(t<0.1)+(A+Bsin(wt))*(t>0.1)
I wrote this off the mind, so you make sure it is continous in time (no gap). Above means that first it ramps the speed up to 0.1s and then after t>0.1 it neglects the first left term and goes to your desired boundary condition.
Nevertheless, without seeing the model it is difficult to guess.
Best,
Danial
Reply | Reply with Quote | Send private message | Report Abuse
July 30, 2010 9:00am UTC in response to Danial Tz
Re: sinosoidal inlet boundary condition
Hi Danial, I found sth.It is not because of the sin function ,it is because of outlet pressure,when I set the outlet pressure to zero it converges but when I set it as 4140 Pa I got that error.where is problem?I want the outlet pressure to be 4140 Pa...
whould u please help me..
Reply | Reply with Quote | Send private message | Report Abuse
July 31, 2010 4:31am UTC in response to maral hassani
Re: sinosoidal inlet boundary condition
hi Danial,
Im waiting for answer..please
thanx
Reply | Reply with Quote | Send private message | Report Abuse
July 31, 2010 9:05am UTC in response to maral hassani
Re: sinosoidal inlet boundary condition
Hi Danial, I found sth.It is not because of the sin function ,it is because of outlet pressure,when I set the outlet pressure to zero it converges but when I set it as 4140 Pa I got that error.where is problem?I want the outlet pressure to be 4140 Pa...
whould u please help me..
Hm, I am not sure why it diverges for you. Well, setting boundary conditions for Navier-stokes is not a trivial task anyway. I tried a set pressure with flow around a cylinder and worked quite well. For N-S the pressure is itself is not important, but the pressure difference. So, with a set inlet velocity profile and a set pressure outlet you should be able to solve the problem. Of course, except that your assumptions for boundary conditions are wrong.
Sorry I could not be of more help.
There are some blood examples in Comsol 3.5a. Maybe you take a look at them too.
Reply | Reply with Quote | Send private message | Report Abuse
July 31, 2010 5:27pm UTC in response to Danial Tz
Re: sinosoidal inlet boundary condition
ok thanx for reply..
Reply | Reply with Quote | Send private message | Report Abuse
Rules and guidelines
