[Nek5000-users] **ERROR**: Failed in HMHOLTZ (axisymmetric)

[nek5000-users at lists.mcs.anl.gov]

Not sure if your convergence problem is related to a stability issue. What's the maximum CFL in this run?
To be stable you need a max{CFL}<0.5-0.6 (assuming TORDER=3) otherwise you have to reduce DT or turn on IFCHAR (OIFS). 

 
-----Original message-----
> From:nek5000-users at lists.mcs.anl.gov <nek5000-users at lists.mcs.anl.gov>
> Sent: Thursday 23rd February 2017 14:51
> To: nek5000-users at lists.mcs.anl.gov
> Subject: [Nek5000-users] **ERROR**: Failed in HMHOLTZ (axisymmetric)
> 
> Dear Neks,
> 
> I have run into this convergence error in PRES while testing an 
> axisymmetric case (**ERROR**: Failed in HMHOLTZ...). See below:
> 
> =============================================================================================
> Step  51118, t= 5.1118000E+01, DT= 1.0000000E-03, C=  0.081 4.9673E+02 
> 5.6732E-03
>               Solving for heat
>   Solving for fluid T T T
>        51118    Hmholtz TEMP:      1   3.1921E-17   2.8762E-14   
> 5.0000E+00
>        51118   5.1118E+01  2.6393E-04 Heat done
>     51118   7.62673E-01   2.44737E-03   3.11629E+02   8 PRES   PROJECT
>   51118   100 **ERROR**: Failed in HMHOLTZ: PRES   1.0006E-04   
> 3.9093E-04   1.0000E-04
>        51118    Hmholtz VELX:      3   8.2217E-09   4.2588E-02   
> 1.0000E-06
>        51118    Hmholtz VELY:      3   1.0309E-08   3.8505E-02   
> 1.0000E-06
>                 L1/L2 DIV(V)    :  -1.5406E-09   2.6956E-03
>                 L1/L2 QTL       :   0.0000E+00   0.0000E+00
>                 L1/L2 DIV(V)-QTL:  -1.5406E-09   2.6956E-03
>        51118   5.1118E+01  8.4209E-03 Fluid done
>      51118  5.1118E+01  4.0915E-03  5.2924E+00 ubar
>   dump history points
>   done :: dump history points
> =============================================================================================
> 
> Time step is so small due to the domain's substantial changes in size, 
> which make the CFL go over 1 and crash otherwise. Velocity is 
> oscillating
>   periodically so CFL will take very low values (such as the one above) 
> and then reach higher values overtime (0.4 or 0.5). Even raising 
> tolerances
> past a reasonable extreme will not make it converge. Any hints?
> 
> Best regards,
> 
> MViturro
> 
> 
> 
> 
> 
> 
> -- 
> MAURO VITURRO BALUFO
> Universidad Politecnica de Madrid
> 
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>