[FMDenaro]

Originally Posted by mihirmakwana6

I found the bug in my code. Actually I was modifying the source term in the poisson solver which actually needs to be fixed till the pressure loop runs. Now, My code runs well and I get my results quite fast.

Also, if I use

P[fixedPoint] = P_fixedValue

than it takes more time then the case when i don't use it.

Thankyou Filippo Maria Denaro sir and Michael Prinkey sir for all your replies

- Mihir

ok, have you checked that you get convergence even without fixing the value? That would be the final guarantee you have the correct source term in the Poisson equation

[mihirmakwana6]

Originally Posted by FMDenaro

ok, have you checked that you get convergence even without fixing the value? That would be the final guarantee you have the correct source term in the Poisson equation

1) yes sir. If i did NOT fix the value at scalar grid point for pressure , then also the solution converges.




2) Also, I am using the entire value of optimum omega for my code as I am using SOR. I plotted graphs for number of Iterations the PRESSURE solver takes for both the RK2 steps. The number of iterations decrease as the solution reaches steady state.

However, I found that just before the STEADY STATE, the number of iterations for both the RK2 steps fluctuate.

I mean, say i get, for the last 6 NUMBER of iterations of the poisson solver

for step 1 : a b a b a b ( where a , b are some integers )

for step 2 : c d c d c d ( where a , b are some integers )

What can be the possible reason for this?

[FMDenaro]

Originally Posted by mihirmakwana6

1) yes sir. If i did NOT fix the value at scalar grid point for pressure , then also the solution converges.




2) Also, I am using the entire value of optimum omega for my code as I am using SOR. I plotted graphs for number of Iterations the PRESSURE solver takes for both the RK2 steps. The number of iterations decrease as the solution reaches steady state.

However, I found that just before the STEADY STATE, the number of iterations for both the RK2 steps fluctuate.

I mean, say i get, for the last 6 NUMBER of iterations of the poisson solver

for step 1 : a b a b a b ( where a , b are some integers )

for step 2 : c d c d c d ( where a , b are some integers )

What can be the possible reason for this?

when you approach the steady state you should still decrease your threshold for convergece....that appens because the guess field used to start will be very close to the correct solution and your solver does to many iterations to correct

[mihirmakwana6]

Originally Posted by FMDenaro

when you approach the steady state you should still decrease your threshold for convergece....that appens because the guess field used to start will be very close to the correct solution and your solver does to many iterations to correct

sir I did not get it.