Storm Sewer Design + Tailwaters

[mbherlihy]

When designing for a 25-year storm network, which discharges into a pond, should we set the tailwater of the the discharging headwall, at the 25 year water surface elevation? My problem occurs when using our design software. (Terramodel)When designing our storm sewer network Terramodel allows us to enter a tail water condition,... and....also assign drainage areas to each inlet. I think I am double counting the amount of rain, b/c common practice in our company, is to enter the 5 minute intensity (8.37 in/hr)of for each inlet, the drainage area and C value. My HGL is going out of the structures by alot..... My thoughts may be that at the 25-year tailwater time (15 hours) the intensity is 0.40 in/hr. Should I be using 0.40 in/hr for data entry into my storm network, when also using the 25 year tailwater?

 

[cvg]

what happens if you have a 25-year storm and the pond is already full? I would assume full pond or at least partly full pond to be a conservative and reasonable assumption.

 

[jgailla]

When we design a drainage system for a certain storm, we use the calculated high water level in the pond for that storm. This is good for reviewers as they can immediately check the tailwater for the storm. It is slightly conservative, as the high water level in the pond is likely to occur after the peak flow through the inlets. Justification for using a lower tailwater level would be difficult. We usually design storm sewers for a 5 or 10 year storm and check that the calculated HGL starting from the high pond level during that storm event is no higher than the grate. A 25 year storm will be hard to design a sewer system to. You will end up with very large pipes, raising the inlets, or both. You should be using a short intensity rainfall for the pipe sizing in any case. We usually use 10 minutes as a minimum. If you design the pipes for the 0.4 in/hr instead of 8.37 in/hr, your drainage system will flood at the grates in even a low intensity storm. I would suggest reading your state DOT drainage manual, which should be available on the Internet.

 

[mbherlihy]

I think i understand most of what you said for the design and sizing of the pipe, use the 5 or 10 minute intensity to generate your HGL and inlet inflows. However,... at the outlet of the HGL, to say that for inflows at the 5 or 10 minute time, your Pond level is at the 25 year high water mark seems incorrect. This highwater mark in our modeling software occurs 15 hours later. Seems like we should either use the 5/10minute Tc for storm design, and have a 5/10 minute tailwater....or use the tailwater at the 25 year level, but then show the inflows for the 15th hour intensity. I hope I don't sound too confused...i generally follow your reasoning, when sizing pipes, we want to make sure there is capacity at the inlet and int the pipe, therefore use the low Tc, but when generating a HGL witha tailwater,...there lies my problem. Thoughts?

 

[bltseattle]

mbherlihy,
It seems to me that you are presuming that you have a "design storm" type event with a single peak. In reality storms often have several peaks, so there could be another intense 5-minute downpour at some time after the initial peak. It is difficult to get a handle on without a long-term rainfall/hydrologic analysis, however, so it is simpler to take the conservative approach and design for pond-full tailwater.

An adjustment that could be made if you have a long pipe system or a very flat system, however, would be to look at the travel time of water between the most upstream inlet to the pond and use the rainfall intensity that corresponds to that duration. Most times piped storm drain systems on small to medium sites (say 0.1 to 10 acres) are going to be designed using a 5- or 10-minute intensity.

As jgaila says, 0.40 in/hr intensity seems too low for designing a pipe system. Contrasting what jgaila says, for our less-intense climate in the Pacific northwest, we often design piped conveyances for 25-yr and occasionally 100-yr - it all depends on the regulating agencies design criteria. If they don't have any, then a 10- to 25-yr is what we would use, depending on what is at risks when the capacity is exceeded.

 

[cvg]

you should generally take the conservative route for drainage design - flooding after the system is constructed does not make the engineer of record look good.

It seems you are mixing apples and oranges. You are using the rational method to size your storm drain for the peak flow. However, you cannot introduce a unit hydrograph method into the design - rational time of concentration and hydrograph peaks have no relation to one another.

bltseattle is correct in that your design storm may have more than one peak. In addition, the peak flow that your storm drain is designed for does not necessarily occur 10 minutes after it starts raining. It could happen at any time. And unless your pond drains very quickly, you could have back to back storms which would leave your pond partially full before your 25-year rainfall even begins.