HEC-RAS ConSpan Modeling Help

[jvfinch]

I am modeling a proposed ConSpan culvert crossing in HEC-RAS 3.1.2. I have input a 32' span by 8' rise ConSpan culvert into the HEC-RAS bridge editor (i.e. I input the coordinates of the ConSpan in the bridge section, not the culvert section). I setup the ineffective flow, etc. for the 32' x 8' ConSpan. I was able to obtain a run with some reasonable/sensible results. The ConSpan invert elevation is 478.89. I obtained a water surface elevation at section #3 (i.e. just upstream of bridge) of 488.56. I then input a 32' span by 10' rise ConSpan in attempt to decrease the upstream water surface elevation and reran the model. The calculated water surface elevation at the same section #3 upstream of the bridge is now 489.93. In both situations, the culvert opening is submerged, so I selected the high flow method of pressure/weir to model the culvert. I cannot for the life of me figure out why when I increased the flow area/culvert size (i.e. went from a 32' x 8' culvert to a 32' x 10' culvert) I obtained a higher water surface elevation. I would expect that as you increase the culvert size, you would decrease the upstream water surface elevation. I originally thought that maybe the 32' x 8' span was submerged, meaning pressure method was used, and the 32' x 10' span was unsubmerged, meaning energy method was used. However, this is not the case because in both situations, the culverts are submerged. Has anyone ever obtained weird results like this or can anyone point me in the right direction to fix the problem. Any help with this situation would be greatly appreciated.

 

[gbam]

OK,
Did anything else change like roadway elevations/weir elevations as a result of the 2 foot increase. Is there combined flow, overtopping & culvert flow? Have you checked inlet vs outlet control? Also, did you try using the cuvert routine for both situations? All I can really say is to verify your input for both situations. I have typically checked my results using HDS 5 or some external method to see if the changes are odd.

Good Luck debugging.

 

[jvfinch]

Absolutely nothing changed. All I did was literally cut and paste the new coordinates for the new rise height, and then hit compute. The roadway does not even come close to overtopping in either scenario. Both are under pressurized flow, basically acting as sluice gates. I am modeling as a bridge, so there is no inlet/outlet control check. The reason I am modeling as a bridge is because we are not touching the channel, just simply spanning over it. Therefore, I would not model this as a culvert because during low flow events, the flow will stay confined to the main channel and wouldn't even see the ConSpan.

Could the problem lie somewhere within the way HEC-RAS models bridges? I did notice that the velocities are approx. 2 ft/sec higher for the smaller culvert at the section just inside the bridge than for the larger culvert. Is it possible that the floodplain is trying to squeeze through a smaller culvert, thus increasing velocities enough to make the water surface actually be lower than the situation with the larger culvert? Maybe the larger culvert allows more flow area for the floodplain, which in turn decreases the velocities near the bridge, thus causing the water surface elevation to be higher upstream of the bridge. I don't know, just talking out of desperation.

Anyway, I have checked my input data for days now. I have consulted with numerous experienced HEC-RAS users who looked over my model and say they cannot figure it out either. I am running out of time and options. Anyone who knows anything I could try, please help!!!!!!!!

 

[gbam]

Hmmm,
Not sure either. The change in velocity head is only 0.06 feet. Has the approach velocity changed upstream of the structure? I have not encountered this, yet. In both runs do the bridge losses show the difference and is the same method used ? That may explain your question about bridge modeling. I cannot remember but I think you can the bridge analysis to be the same for both to see if you get different results. Its been a few years since I had to model a bridge. I understand the time sensitivity, if you can, try looking at an array of heights holding the span constant to see if you can see a relationship for the structures.

Other than that, I am not sure what else to do.

 

[jvfinch]

For both model runs, the pressure method was selected as the bridge modeling methodology since the culvert is submerged but does not overtop the roadway. The velocities are higher for the smaller culvert in the vicinity of the bridge. I guess this has to be due to the flow trying to squeeze through a smaller opening. I can't imagine though that this increase in velocity would cause a 1.07' difference in the water surface elevation. I really have no idea how to proceed with this. Obviously, I will not be able to explain to the client that a larger culvert backs up more water, and I am running out of time. I am going to continue fooling with model inputs (i.e. various culvert sizes) and calculation methodologies and see if I can figure anything out. If anyone knows of any reason why this is happening, or has any sort of suggestion, please post it.

Thanks for you help.