Excessive 1D weir (W) channel headloss when submerged

[DanielCopelin 0]

Hi,

 

I have a weir (W) channel with a crest level of 19m as part of a 1D network. The weir geometry is being read from a mid cross-section file, and the check files seem to indicate that the geometry is being read correctly and the hydraulic properties seem correct.

 

However, I am seeing very large afflux (approx 1.5m) across the structure even under very high flows where the weir is drowned by ~20m+ of water over the crest. By comparison, a steady state HECRAS model only predicts about 0.15m difference in upstream and downstream water level, which is more reasonable. If I model the W channels as normal S channels, I get results that are similar to HECRAS.

 

I have a suspicion that the W channels are not using the submerged flow regime, but I don't know how to go about diagnosing this. I can't seem to find anything of relevance in the check files and .eof file.

 

I am using the latest version of TUFLOW.

 

Does anyone have any experience with this?

[katebradbrook 0]

Daniel,

Yes, I had noticed this problem and did a considerable amount of benchmarking of weirs modelling between ESTRY, ISIS, HEC-RAS and hand-calculations and contacted Bill Syme.  He investigated and recommended using Weir Flow == Method A in these situations where weirs are drowned and we are not expecting much headloss....see if this helps your weir....

all the best,

Kate

[DanielCopelin 0]

Daniel,

Yes, I had noticed this problem and did a considerable amount of benchmarking of weirs modelling between ESTRY, ISIS, HEC-RAS and hand-calculations and contacted Bill Syme.  He investigated and recommended using Weir Flow == Method A in these situations where weirs are drowned and we are not expecting much headloss....see if this helps your weir....

all the best,

Kate

 

Thanks, Kate! This seems to have worked. I have upstream levels within 0.1m of HECRAS now.

[Bill Syme 0]

Hi Daniel and thanks Kate.

 

For further explanation, Method B (the default in recent releases), was introduced to provide better stability for weirs embedded within 2D domains (eg. over culverts or a bridge that are modelled in 2D using 2D FCs).  We're still investigating, however, it seems that in some situations it does not converge well and can cause a large head drop.  For the 2013 release, you'll be able to select which method to apply on a weir by weir basis, and Method A is likely to be set back as the default (unless using a Defaults == Pre... command).  In the meantime, should there be a concern over the head drop of a drowned weir, try setting Weir Flow == Method A in the .ecf file.  If anyone has any feedback, please add to this topic or email support@tuflow.com.

 

Cheers

Bill