# 5.7.1 solver doing max coeff loops.

 Register Blogs Members List Search Today's Posts Mark Forums Read

 July 1, 2005, 12:50 5.7.1 solver doing max coeff loops. #1 Stevie Wonder Guest   Posts: n/a Hello people, Have you ever seen a case that the residuals (U,V,W,P) converge in the inner coefficent loops but the solver keeps doing the maximum set coeff. loops for every time step in a transient simulation? Details of the ccl is given below:

SOLVER CONTROL:

Blend Factor = 0.75

Option = Specified Blend Factor

END

CONVERGENCE CONTROL:

Maximum Number of Coefficient Loops = 5

END

CONVERGENCE CRITERIA:

Residual Target = 5e-05

Residual Type = RMS

END

TRANSIENT SCHEME:

Option = Second Order Backward Euler

END

END As stated, I have defaults definitions (i.e. no expert parameters) but the solver keeps solving all 5 coeff. loops as can be seen in the output presented.

============================================= =========================

TIME STEP = 230 SIMULATION TIME = 2.12E-01 CPU SECONDS = 8.31E+05

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 1

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 1.02 | 1.9E-04 | 3.7E-03 | 1.6E-02 OK|

| V-Mom | 1.00 | 2.0E-04 | 4.0E-03 | 5.5E-03 OK|

| W-Mom | 1.00 | 1.9E-04 | 3.7E-03 | 1.7E-02 OK|

| P-Mass | 0.92 | 1.0E-05 | 1.1E-03 | 24.3 5.4E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 0.67 | 5.2E-04 | 3.5E-02 | 10.9 7.3E-02 OK|

| vv-RS | 1.02 | 4.0E-04 | 1.4E-02 | 10.9 6.8E-02 OK|

| ww-RS | 0.79 | 8.0E-04 | 1.4E-01 | 5.9 6.8E-02 OK|

| uv-RS | 0.91 | 4.7E-04 | 1.5E-02 | 15.8 4.9E-02 OK|

| uw-RS | 1.17 | 1.2E-03 | 3.2E-02 | 10.9 7.4E-02 OK|

| vw-RS | 1.15 | 4.6E-04 | 1.6E-02 | 15.8 4.2E-02 OK|

| E-Diss.K | 1.49 | 7.1E-04 | 6.0E-02 | 12.3 2.6E-03 OK|

+----------------------+------+---------+---------+------------------+

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 2 CPU SECONDS = 8.32E+05

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.52 | 1.0E-04 | 1.7E-03 | 1.3E-01 ok|

| V-Mom | 0.48 | 9.7E-05 | 3.0E-03 | 1.0E-01 ok|

| W-Mom | 0.53 | 1.0E-04 | 1.8E-03 | 1.3E-01 ok|

| P-Mass | 1.57 | 1.6E-05 | 1.3E-03 | 10.3 4.8E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 3.00 | 1.6E-03 | 3.9E-02 | 10.9 4.2E-02 OK|

| vv-RS | 0.87 | 3.5E-04 | 1.0E-02 | 10.9 8.1E-02 OK|

| ww-RS | 1.67 | 1.3E-03 | 1.0E-01 | 10.9 6.9E-02 OK|

| uv-RS | 1.33 | 6.2E-04 | 2.0E-02 | 10.9 8.4E-02 OK|

| uw-RS | 0.27 | 3.3E-04 | 1.9E-02 | 10.9 8.1E-02 OK|

| vw-RS | 1.22 | 5.6E-04 | 1.8E-02 | 10.9 9.1E-02 OK|

| E-Diss.K | 0.80 | 5.7E-04 | 8.4E-02 | 12.3 7.3E-03 OK|

+----------------------+------+---------+---------+------------------+

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 3 CPU SECONDS = 8.32E+05

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.57 | 5.7E-05 | 1.2E-03 | 1.8E-01 ok|

| V-Mom | 0.55 | 5.4E-05 | 1.6E-03 | 1.3E-01 ok|

| W-Mom | 0.57 | 5.7E-05 | 1.2E-03 | 1.8E-01 ok|

| P-Mass | 1.13 | 1.9E-05 | 1.5E-03 | 10.3 2.0E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 0.45 | 7.0E-04 | 2.7E-02 | 10.9 4.7E-02 OK|

| vv-RS | 1.19 | 4.2E-04 | 2.0E-02 | 10.9 6.5E-02 OK|

| ww-RS | 1.08 | 1.4E-03 | 3.0E-01 | 5.9 9.6E-02 OK|

| uv-RS | 0.82 | 5.1E-04 | 1.7E-02 | 10.9 9.5E-02 OK|

| uw-RS | 3.61 | 1.2E-03 | 3.5E-02 | 10.9 7.1E-02 OK|

| vw-RS | 0.90 | 5.1E-04 | 1.8E-02 | 10.9 8.4E-02 OK|

| E-Diss.K | 0.42 | 2.4E-04 | 3.6E-02 | 12.3 3.6E-03 OK|

+----------------------+------+---------+---------+------------------+

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 4 CPU SECONDS = 8.33E+05

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.78 | 4.4E-05 | 1.5E-03 | 1.7E-01 ok|

| V-Mom | 0.73 | 3.9E-05 | 1.1E-03 | 1.3E-01 ok|

| W-Mom | 0.78 | 4.4E-05 | 1.2E-03 | 1.6E-01 ok|

| P-Mass | 0.54 | 1.0E-05 | 6.1E-04 | 10.3 3.2E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 1.35 | 9.4E-04 | 2.6E-02 | 10.9 6.9E-02 OK|

| vv-RS | 0.91 | 3.8E-04 | 1.2E-02 | 10.9 8.0E-02 OK|

| ww-RS | 0.85 | 1.2E-03 | 2.1E-01 | 10.9 6.1E-02 OK|

| uv-RS | 1.11 | 5.7E-04 | 1.9E-02 | 10.9 8.4E-02 OK|

| uw-RS | 0.41 | 4.8E-04 | 2.0E-02 | 10.9 7.9E-02 OK|

| vw-RS | 0.94 | 4.8E-04 | 1.5E-02 | 10.9 8.9E-02 OK|

| E-Diss.K | 1.06 | 2.6E-04 | 4.0E-02 | 12.3 6.2E-03 OK|

+----------------------+------+---------+---------+------------------+

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 5 CPU SECONDS = 8.34E+05

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.78 | 3.4E-05 | 1.2E-03 | 1.6E-01 ok|

| V-Mom | 0.78 | 3.1E-05 | 7.9E-04 | 1.2E-01 ok|

| W-Mom | 0.82 | 3.6E-05 | 1.4E-03 | 1.6E-01 ok|

| P-Mass | 0.95 | 9.5E-06 | 1.1E-03 | 10.3 4.3E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 0.65 | 6.1E-04 | 3.2E-02 | 10.9 6.5E-02 OK|

| vv-RS | 0.90 | 3.4E-04 | 1.5E-02 | 10.9 5.4E-02 OK|

| ww-RS | 0.60 | 7.3E-04 | 2.3E-02 | 5.9 9.9E-02 OK|

| uv-RS | 0.65 | 3.7E-04 | 1.2E-02 | 10.9 8.7E-02 OK|

| uw-RS | 1.97 | 9.4E-04 | 2.6E-02 | 10.9 5.9E-02 OK|

| vw-RS | 0.93 | 4.4E-04 | 1.8E-02 | 10.9 8.2E-02 OK|

| E-Diss.K | 0.82 | 2.1E-04 | 2.6E-02 | 12.3 5.9E-03 OK|

+----------------------+------+---------+---------+------------------+

============================================= =========================

TIME STEP = 231 SIMULATION TIME = 2.13E-01 CPU SECONDS = 8.34E+05

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 1

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.99 | 1.9E-04 | 3.6E-03 | 1.6E-02 OK|

| V-Mom | 0.99 | 2.0E-04 | 4.0E-03 | 5.4E-03 OK|

| W-Mom | 0.99 | 1.9E-04 | 3.6E-03 | 1.6E-02 OK|

| P-Mass | 0.76 | 8.0E-06 | 7.9E-04 | 24.3 5.6E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 2.10 | 1.1E-03 | 3.2E-02 | 10.9 5.3E-02 OK|

| vv-RS | 0.85 | 3.4E-04 | 1.1E-02 | 10.9 6.6E-02 OK|

| ww-RS | 0.98 | 7.8E-04 | 7.9E-02 | 10.9 5.0E-02 OK|

| uv-RS | 1.20 | 5.7E-04 | 2.0E-02 | 10.9 8.9E-02 OK|

| uw-RS | 0.78 | 9.4E-04 | 3.2E-02 | 10.9 6.1E-02 OK|

| vw-RS | 0.94 | 4.3E-04 | 1.3E-02 | 10.9 8.9E-02 OK|

| E-Diss.K | 1.03 | 7.3E-04 | 8.1E-02 | 12.3 3.2E-03 OK|

+----------------------+------+---------+---------+------------------+

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 2 CPU SECONDS = 8.35E+05

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.52 | 9.7E-05 | 1.9E-03 | 1.3E-01 ok|

| V-Mom | 0.48 | 9.6E-05 | 3.0E-03 | 1.0E-01 ok|

| W-Mom | 0.51 | 9.6E-05 | 1.7E-03 | 1.3E-01 ok|

| P-Mass | 2.05 | 1.6E-05 | 1.4E-03 | 10.3 4.7E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 1.23 | 1.3E-03 | 4.5E-02 | 10.9 3.2E-02 OK|

| vv-RS | 0.95 | 3.3E-04 | 1.6E-02 | 10.9 6.7E-02 OK|

| ww-RS | 1.80 | 1.4E-03 | 1.8E-01 | 10.9 5.4E-02 OK|

| uv-RS | 0.85 | 4.8E-04 | 1.3E-02 | 10.9 7.9E-02 OK|

| uw-RS | 0.97 | 9.0E-04 | 2.9E-02 | 10.9 6.1E-02 OK|

| vw-RS | 1.31 | 5.6E-04 | 2.3E-02 | 10.9 7.9E-02 OK|

| E-Diss.K | 0.60 | 4.4E-04 | 7.3E-02 | 12.3 5.9E-03 OK|

+----------------------+------+---------+---------+------------------+

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 3 CPU SECONDS = 8.36E+05

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.58 | 5.6E-05 | 1.4E-03 | 1.8E-01 ok|

| V-Mom | 0.56 | 5.4E-05 | 1.3E-03 | 1.3E-01 ok|

| W-Mom | 0.59 | 5.7E-05 | 1.1E-03 | 1.7E-01 ok|

| P-Mass | 1.00 | 1.6E-05 | 9.6E-04 | 10.3 1.9E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 0.20 | 2.7E-04 | 1.9E-02 | 10.9 9.1E-02 OK|

| vv-RS | 1.29 | 4.2E-04 | 2.1E-02 | 10.9 7.4E-02 OK|

| ww-RS | 0.45 | 6.4E-04 | 1.4E-01 | 10.9 7.2E-02 OK|

| uv-RS | 1.09 | 5.3E-04 | 2.2E-02 | 10.9 9.2E-02 OK|

| uw-RS | 0.96 | 8.7E-04 | 3.1E-02 | 10.9 7.9E-02 OK|

| vw-RS | 0.75 | 4.2E-04 | 1.3E-02 | 15.9 4.1E-02 OK|

| E-Diss.K | 0.63 | 2.8E-04 | 2.6E-02 | 12.3 3.9E-03 OK|

+----------------------+------+---------+---------+------------------+

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 4 CPU SECONDS = 8.37E+05

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.75 | 4.2E-05 | 9.9E-04 | 1.8E-01 ok|

| V-Mom | 0.72 | 3.9E-05 | 8.0E-04 | 1.3E-01 ok|

| W-Mom | 0.75 | 4.3E-05 | 1.4E-03 | 1.6E-01 ok|

| P-Mass | 0.61 | 9.9E-06 | 8.3E-04 | 10.3 3.0E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 4.89 | 1.3E-03 | 4.9E-02 | 10.9 5.0E-02 OK|

| vv-RS | 0.84 | 3.6E-04 | 1.7E-02 | 10.9 5.9E-02 OK|

| ww-RS | 1.93 | 1.2E-03 | 4.1E-02 | 10.9 5.3E-02 OK|

| uv-RS | 0.85 | 4.5E-04 | 1.2E-02 | 10.9 7.0E-02 OK|

| uw-RS | 0.60 | 5.2E-04 | 1.6E-02 | 10.9 5.9E-02 OK|

| vw-RS | 1.28 | 5.4E-04 | 2.3E-02 | 10.9 6.7E-02 OK|

| E-Diss.K | 1.00 | 2.8E-04 | 4.4E-02 | 12.3 3.8E-03 OK|

+----------------------+------+---------+---------+------------------+

----------------------------------------------------------------------

COEFFICIENT LOOP ITERATION = 5 CPU SECONDS = 8.37E+05

----------------------------------------------------------------------

| Equation | Rate | RMS Res | Max Res | Linear Solution |

+----------------------+------+---------+---------+------------------+

| U-Mom | 0.79 | 3.3E-05 | 1.3E-03 | 1.6E-01 ok|

| V-Mom | 0.74 | 2.9E-05 | 7.9E-04 | 1.2E-01 ok|

| W-Mom | 0.75 | 3.2E-05 | 9.0E-04 | 1.5E-01 ok|

| P-Mass | 0.87 | 8.6E-06 | 1.1E-03 | 10.3 4.0E-02 OK|

+----------------------+------+---------+---------+------------------+

| uu-RS | 0.26 | 3.5E-04 | 2.1E-02 | 10.9 6.2E-02 OK|

| vv-RS | 0.96 | 3.4E-04 | 1.7E-02 | 10.9 6.2E-02 OK|

| ww-RS | 0.22 | 2.7E-04 | 1.4E-02 | 10.9 9.3E-02 OK|

| uv-RS | 1.15 | 5.2E-04 | 2.3E-02 | 10.9 7.3E-02 OK|

| uw-RS | 1.91 | 1.0E-03 | 3.3E-02 | 10.9 6.4E-02 OK|

| vw-RS | 0.65 | 3.5E-04 | 1.1E-02 | 10.9 9.6E-02 OK|

| E-Diss.K | 0.65 | 1.8E-04 | 1.8E-02 | 12.3 7.4E-03 OK|

 July 1, 2005, 20:49 Re: 5.7.1 solver doing max coeff loops. #2 Jeff Guest   Posts: n/a S.W. The solver keeps solving coefficient loops until "all" variables have met the convergence criteria. Your convergence criteria is set to 1.0E-5. On the iterations you posted, Mass (P) is makeing it below this target (i.e. 9.xE-6) but none of the other variables are. So, after your prescribed max coefficient loops the solver gives up and moves on to the next time step. Solution: 1. Use a smaller time step 2. Allow more coefficent loops (10 is typical) 3. Raise your convergence criteria Solution 1 is your best bet here. Notice the lower case "ok" on U,V,W-momentum. This means that the coefficient loops are barely acceptable. You probably need a smaller time step here to get acceptable coefficient loop behavior, and to get all the variables to the target RMS value. Jeff

 July 2, 2005, 10:28 Re: 5.7.1 solver doing max coeff loops. #3 Rui Guest   Posts: n/a Hi, I appologize for being meddling in this topic, but Stevie Wonder's Residual Target RMS is 5E-5 (not 1E-5), and acording to the documentation: "The CFX-Solver will terminate the run when the equation residuals calculated using the method specified are below the Residual Target value. Note that the residual level for turbulence transport equations (for k, epslon, and the Reynolds stress components) does not form part of the convergence criteria." After 4 coefficient loops, the U, V, W and P equations RMS residuals are below 5E-5, and the turbulence equations do not form part of the convergence criteria. The cause why the Solver doesn't jump to the next time step after 4 coefficient loops may be, as Jeff noticed, the lower case "ok" on the U, V and W equations. I would suggest to set the Maximum Number of Coefficient Loops to a high value (50 for example), and start from an intermediate result obtained with Maximum Number of Coefficient Loops = 5. Check if the Solver jumps to the next time step when all the U, V, W and P equations have a capital "OK". Regards, Rui

 July 4, 2005, 08:29 Re: 5.7.1 solver doing max coeff loops. #4 Stevie Wonder Guest   Posts: n/a Hey guys, Thanks for your awesome inputs. ;-) My problem is that I was neglecting the lower case "ok" in the hydrodynamic residual equations. I'm considering in reducing my time step size. I might be wrong but I've heard from someone from CFX that in a single-phase incompressible flow coefficient loops greater than 8 could even be dangerous to your solution. Reducing the time step is best thing applicable in this case. Cheers, S. W.

 July 4, 2005, 19:12 Re: 5.7.1 solver doing max coeff loops. #5 Glenn Horrocks Guest   Posts: n/a Hi, I think saying using more than 8 iterations per timestep is "dangerous to your solution" for single phase is going a bit far. Almost always you will find that it is not optimum and that using a smaller timestep and less iterations per timestep is more efficient. Regards, Glenn

 July 5, 2005, 12:31 Re: 5.7.1 solver doing max coeff loops. #6 Pete Guest   Posts: n/a Hi Glenn, I think sometime you will need extra few timesteps if your initial solution is not good enough and you're using transient simulation to check if the flow is unsteady in nature.... Let me know if I'm wrong.. Cheers.

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post Peter_600 OpenFOAM 4 August 2, 2014 09:52 ozzythewise OpenFOAM Running, Solving & CFD 3 February 8, 2011 16:28 Roued CFX 1 October 2, 2001 16:49 tokai CFX 10 July 17, 2001 16:25 cfd guy CFX 4 May 8, 2001 06:04

All times are GMT -4. The time now is 18:49.