Half-Moon Baffle in culvert calculation

[DMcGrath]

I have a 48" culvert with a steel plate installed in the lower half that blocks flow until the water level is at least two feet deep before it would overflow this plate and enter the culvert. How would I go about calculating a rating table for this culvert? I don't see an option for anything like this in HY-8, and building a HEC-RAS model seems to be the sledgehammer approach.

 

[Ryb01]

Hi DMcGrath,

From a glance at the photo you've attached, I'm not sure that there is a simple solution to your question. There's a number of factors that are put in play with this scenario depending on the flow conditions you are experiencing. I'm also not sure that the sole use of software would be the best approach for this type of application. I'll explain my rationale in detail a little further down.

It looks like the restrictor plate has been used at the culvert entrance to utilize effective head produced overtop of the catchbasin infront of the culvert. Possibly for minor system flows and to trap sediment from entering the culvert?? The rating curve for this catchbasin structure would have to be taken into consideration in conjunction with the culvert analysis as you would have a spilt flow condition between the CB and the restrictor plate.

The initial portion of the rating curve would only contain the catchbasin capture up to the weir crest elevation. From the crest of the restrictor plate to the obvert of the pipe weir flow would govern and finally from the obvert up, orifice flow would govern. All of which would be used to compile your rating curve.

The next considerations is where things get interesting.

The standard sharp crested weir equations take into account either vertical or sloped walls. Trapezoidal or cipoletti type weir equations account for an increasing flow area as head increases. However, in this case the effective flow area overtop of the weir decreases as head increases and as the headwater approaches the obvert of the pipe. With this in mind it may be advantageous to analyze the flow over the restrictor plate in flow slices, assuming a vertical wall wier at say 1" or 2" inch increments over the restrictor plate, adding each increment to obtain a total flow at a particular depth. The smaller the increments, the more accurate the flow.

Once flow touches the obvert of the pipe, this becomes a pressure condition and accordningly an orifice equation should be used.

The steps above only consider a "free flowing condition". Further considerations for outlet control would need to be examined taking into account minor losses and tailwater conditions. If applicable these could also influence your rating curve. This is where the use of a program such as Hec-Ras may be useful.

Given the limitations of HEC-Ras and even say SWMM, there would need to be some back and forth between your standard spreadsheet analysis's and the software approach. This can be a very tedious and time consuming task, depending on the level of detail you require.

There may be a more efficient way to complete this analysis, I'd be open to any further suggestions, however this is how I'd approach the challenge from an outside perspective.

Hope this helps. Good luck.

 

[DamEng1]

The steel plate will initially act as a sharp crested weir and then should transition to an orifice flow above the top of the pipe.

 

[beej67]

If you're looking for a quick approximation, calculate the open area, divide that area by the pipe diameter, and you'll get a "characteristic height" of a little less than D/2. Then pretend the thing is a rectangular weir of a width of 48" and a crown at that "characteristic height." That should give you a slight overestimation of discharge.

If you want to get super fancy, do the same sort of process, but divide the area by (D/2) and get a "characteristic width." Then use that width and a height of D/2 to create another hypothetical rectangular weir which would give you a slight underestimation of discharge. Then develop a "user defined rating curve" for your software program that splits the difference between the two hypothetical weirs.

AFAIK there's no generally accepted way to handle this kind of geometry in engineering, so you'll have to fall back on some assumptions and be very open with your reviewers / client about what you're doing. Document your assumptions.

 

[gbam]

Simple approach would use both the weir and orifice equations. Combine into a controlling headwater rating curve and evaluate as DamEng1 points out.