Establish an initial trial size. The trial size may be picked at random or judiciously, based on experience. However, one convenient rule-of-thumb is to assume inlet control and proceed as follows: Determine the maximum practical rise of culvert (D

max

) and the maximum allowable headwater depth (HW

max

). Determine a trial head using Equation 8-1.

h
= allowable effective head (ft. or m)
HW
max
= allowable headwater depth (ft. or m)
D
max
= maximum conduit rise (ft. or m).

A
= approximate cross-sectional area required (sq.ft. or m
2
)
Q
= design discharge (cfs or m
3
/s).

A properly designed culvert has an effective flow area similar in height and width to the approach channel section so that approach velocities and through-culvert velocities are similar.
For a box culvert, determine the required width, W, as A/D
max
. Round W to the nearest value that yields a whole multiple of standard box widths. Divide W by the largest standard span S for which W is a multiple. This yields the number of barrels, N. At this point, the determination has been made that the initial trial configuration will be N - S D
max
L, where L is the length of the barrel in feet.
For a circular pipe culvert, determine the ratio of area required to maximum barrel area according to Equation 8-3.

Equation 8-3.
Round this value to the nearest whole number to get the required number of barrels, N. At this point, the determination has been made that the initial trial size culvert will be N - D L circular pipe.
For other shapes, provide an appropriate size such that the cross section area is approximately equal to A.

Determine the design discharge per barrel as Q/N. This assumes that all barrels are of equal size and parallel profiles with the same invert elevations. The computations progress using one barrel with the appropriate apportionment of flow.

Perform a hydraulic analysis of the trial configuration. Generally, a computer program or spreadsheet would be used. Nomographs and simplified hand methods should be used only for preliminary estimates. For the trial configuration determine the inlet control headwater (HW

ic

), the outlet control headwater (HW

oc

) and outlet velocity (v

o

) using Flow Chart A shown in Figure 8‑7. Flow Chart A references Flow Chart B, which is shown in Figure 8‑8.

Evaluate the trial design. At this step in the design process, you have calculated a headwater and outlet velocity for the design discharge through a trial culvert configuration has been calculated.

If the calculated headwater is equal to or is not appreciably lower than the allowable headwater (an indication of culvert efficiency), the design is complete. A good measure of efficiency is to compare the calculated headwater with the culvert depth D. If the headwater is less than the depth, the configuration may not be efficient.
If the calculated headwater is considerably lower than the allowable headwater or lower than the culvert depth D, a more economical configuration may be possible. Choose the trial culvert configuration by reducing the number of barrels, span widths, diameter, or other geometric or material changes. Repeat the calculations from step 2.
If the calculated headwater is equal to or is not appreciably lower than the actual headwater and the culvert is operating as inlet control, an improved inlet may be in order.
If the calculated headwater is greater than the actual headwater, change the trial culvert configuration to increase capacity by adding barrels, widening spans, and increasing diameter. Repeat the calculations from step 2.
If the operation is not inlet control, then the culvert geometry design is complete.
If the culvert is operating with inlet control, the possibility exists for improving the entrance conditions with the aim of reducing the overall cost of the structure. Investigate the design of a flared (or tapered) inlet and associated structure. Because of the cost of the improved inlet, make a careful economic comparison between the design with a normal entrance and the design with an improved inlet.
Check outlet velocities against the predetermined maximum allowable for the site. The culvert for which the calculated headwater is satisfactory may have an excessive outlet velocity. Excessive velocities are usually caused by a steep slope or a computational error. The definition of "excessive" is usually an engineering judgment based on local and soil conditions, but as a general rule, anything over 12 fps is considered excessive.

Develop a hydraulic performance curve using the procedures outlined in the section. An overall hydraulic performance curve for the designed culvert indicates headwater and outlet velocity characteristics for the entire range of discharges. At an absolute minimum, the additional analysis of the 1% AEP discharge is required for both the existing and the proposed conditions.

The design can be completed if the results of the headwater and outlet velocity represent an acceptable risk and conform to FEMA NFIP requirements. (See Chapter 2 and pertinent parts of the
Project Development Policy Manual
for more details.)
However, if any of the hydraulic characteristics are unacceptable, some adjustment to the culvert design may be in order.