Difference between revisions of "TUFLOW Message 3006"
Melissa1987 (talk | contribs) |
Melissa1987 (talk | contribs) |
||
| (One intermediate revision by one other user not shown) | |||
| Line 1: | Line 1: | ||
{{TUFLOW_Message | {{TUFLOW_Message | ||
| − | |tuflow_message=ERROR 3006 | + | |tuflow_message=ERROR 3006 - Timestep Repeats must be >= 0. |
<br> | <br> | ||
| − | |alt_msg= | + | |alt_msg=NA<br> |
|type=[[ERROR]] | |type=[[ERROR]] | ||
| − | |message_desc= | + | |message_desc=When running in adaptive timestep mode (the default) HPC will monitor the the courant, celerity, and peclet (diffusion) control numbers on a timestep by timestep basis. |
| − | + | At the end of a timestep, if NaNs are encountered in the solution, or any of the three control numbers has jumped to more than 20% above their control targets, the step results are | |
| + | invalidated and the step repeated with a reduced timestep. Normally the solution is then able to progress stably again. However, consecutive repeat timesteps indicate a | ||
| + | severe discontinuity at that point. HPC will abandon the solution if the number of consecutive repeat timesteps exceeds a threshold limit. | ||
| + | The default limit is 10, but is able to be adjusted with the "<font color="blue"><tt>Timestep Repeats</tt></font>" tcf command. | ||
| + | |||
| + | |suggestions=Generally, changing the timesteps repeats limit is not the best method to solve the problem. | ||
| + | For models that remain dry for a period of time before receiving a sudden inflow, the model timestep may increase during the dry period requiring a number of | ||
| + | consecutive repeat timesteps at the onset of inflow in order to return the timestep to a more typical value. In this case, use the "<font color="blue"><tt>Timestep Maximum</tt></font>" tcf command to limit the model | ||
| + | timestep during the dry period and hence the number of consectutive repeat timesteps at the onset of inflow. | ||
| + | For models that are encountering instabilities, these can sometimes be solved using the "<font color="blue"><tt>Control Number Factor</tt></font>" tcf command to force the adaptive timestep control to use smaller timesteps. | ||
| + | Otherwise, the user will have to troubleshoot the source of the instability through the checkfiles. | ||
| − | |||
|uplink=[[3xxx_TUFLOW_Messages|3xxx Messages]] | |uplink=[[3xxx_TUFLOW_Messages|3xxx Messages]] | ||
| − | |||
}} | }} | ||
Latest revision as of 10:10, 3 March 2023
TUFLOW Message
ERROR 3006 - Timestep Repeats must be >= 0.
Alternate Message
NA
Message Type
ERROR
Description
When running in adaptive timestep mode (the default) HPC will monitor the the courant, celerity, and peclet (diffusion) control numbers on a timestep by timestep basis.
At the end of a timestep, if NaNs are encountered in the solution, or any of the three control numbers has jumped to more than 20% above their control targets, the step results are
invalidated and the step repeated with a reduced timestep. Normally the solution is then able to progress stably again. However, consecutive repeat timesteps indicate a
severe discontinuity at that point. HPC will abandon the solution if the number of consecutive repeat timesteps exceeds a threshold limit.
The default limit is 10, but is able to be adjusted with the "Timestep Repeats" tcf command.
Suggestions
Generally, changing the timesteps repeats limit is not the best method to solve the problem.
For models that remain dry for a period of time before receiving a sudden inflow, the model timestep may increase during the dry period requiring a number of
consecutive repeat timesteps at the onset of inflow in order to return the timestep to a more typical value. In this case, use the "Timestep Maximum" tcf command to limit the model
timestep during the dry period and hence the number of consectutive repeat timesteps at the onset of inflow.
For models that are encountering instabilities, these can sometimes be solved using the "Control Number Factor" tcf command to force the adaptive timestep control to use smaller timesteps.
Otherwise, the user will have to troubleshoot the source of the instability through the checkfiles.
| Up |
|---|
 3xxx Messages
|