CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   negative mass flow at the outlet of vessel at the last timestep! (http://www.cfd-online.com/Forums/cfx/76204-negative-mass-flow-outlet-vessel-last-timestep.html)

 smn May 18, 2010 03:20

negative mass flow at the outlet of vessel at the last timestep!

hi:
i have modeled a blood vessel with pulsatile velocity inlet and constant pressure outlet.all of the results are thue and there is no problem,only:
at the last timestep (end time) the number of mass flow at the outlet becomes negative...
can anyone help me?
best regards

 ghorrocks May 18, 2010 07:58

What is wrong with a negative flow rate? Sounds perfectly possible to me based on how you have described the setup. If this does not physically happen then I think you will find your boundary condition setup is incorrect.

 smn May 18, 2010 08:27

Quote:
 Originally Posted by ghorrocks (Post 259274) What is wrong with a negative flow rate? Sounds perfectly possible to me based on how you have described the setup. If this does not physically happen then I think you will find your boundary condition setup is incorrect.
mass flow is positive in all timesteps except the last one.vector plot shows that the flow leaves the "outlet boundary" in all of the timesteps also the last one.how can it be possible?
i have tried both "outlet" and "opening" boundary conditions.
i should add that for example there is 164 timesteps and the mass flow at the last timestep is nearly the same as the mass flow of the 163th and this is true in my problem,but the negative sign in fuction calculator in the post cfx!!!!
regards

 ghorrocks May 18, 2010 19:10

Quote:
 how can it be possible?
Because the flow reverses at time step 164. Why do you say it is not possible? What do you know to say it cannot happen? Make sure you do not confuse the positive average flow with a momentary flow reversal. Even if the flow reverses for a short time the average flow is still positive.

 smn May 22, 2010 09:50

hi ghorrocks and thanks for your replies:
i have mentioned that the
1-flow arrows (vector plot)dont show reverse flow.
2-also the blood flow in the artery can not reverse.
3-i have modeled two period of blood flow;times 0 to 0.82s and second one 0.82 to 1.64 s.how is it possible to have reverse flow in the last timestep?i have also tested very small timesteps.
regards

 Claudia May 23, 2010 11:57

If you look to a backup, e.g. iteration 150, with CFX post, the mass flow at outlet is positive?

Greets
Claudia

 ghorrocks May 23, 2010 18:54

smn - you have not answered my question. Why can't the flow reverse for an instant? How do you know it does not? Do you have a flow versus time measurement with enough resolution to prove there is no reverse flow?

Just because the time average flow is always forwards does not mean for a small fraction of time reverse flow exists. Often the most interesting results are ones which you did not expect. If you dismiss them as impossible without thinking about it then you have just missed an important result.

 azizireza October 28, 2014 13:47

mass flow sign convention

I answered this question, because recently I have encountered this problem.
After finding negative mass flow on output, I have checked the CFX Reference Guide page 158 (version 12.1) and find mass flow sign convention ssubject which helped me.

 All times are GMT -4. The time now is 08:26.