Culvert 600′ long – use lids?

[Katie]

We need to size a box culvert for a canal. The ‘culvert’ is 600 feet long.

Does this take it out of the realm of ‘culvert’ and put it into pressurized pipe simply due to the length, or would it depend on the size of the culvert?

Would I use lidded cross sections to represent this?

I’ve read through other threads on the forum about long culverts and read in the manuals about lidded cross sections, but it seemed that lidded cross sections assume open channel flow and may not fully account for entrance/exit losses like the culvert routines do. Do they include these losses?

We are going for zero rise in WSE after the culvert is installed, so I’m guessing that may lead to a large culvert that doesn’t fill completely, which I assume would mean the lidded cross sections would be OK. But if the culvert is smaller, would lidded cross sections still be appropriate?

Would there be a need or benefit to running this as unsteady?

[Anonymous]

The short answer:

To get a zero rise WSEL, the culvert is going to have to be “channel sized” with almost no contraction as the flow goes into the culvert. Either a culvert or cross sections with lids should work ok. If you go with a culvert, you should probably set the inlet loss coefficient to “0.0”. Since it really is a culvert, using a culvert gives you culvert specific output and it may be easier than trying “to explain” why cross sections with lids is being used.

The long answer:

If the culvert stays a constant size and stays a constant slope (that is, it doesn’t have a vertical bend), then I think the culvert routines should work just fine as far as the 600 foot length is concerned (whether it is pressurized or not).

Lidded cross sections do not have to be open channel flow. If lidded cross sections go into pressure flow and you are running unsteady, then there is some chance for stability problems. But as long as the run is stable, the results should be ok. The Priessmann’s slot option is pretty much required for unsteady lids that goes pressurized in order to get stability.

Cross sections with lids will handle exit losses just fine. The exit losses will be computed based on the “expansion loss” that is set. The default expansion loss is 0.3. This may need bumping up for the exit losses. This can be set from Geometric Editor, Tables, Contraction…

The inlet is a litte more complicated. A culvert can flow under “inlet” control meaning the size and shape of the inlet is the controlling section and it is not dependent on the length or slope of the culvert nor on the tailwater at the outlet. Inlet control does not happen that often, but it could. Even if you go with cross section with lids, you could enter a culvert just to see what the inlet control headwater answer is. As long as this inlet control answer is lower than the cross sections with lids answer (immediately upstream of the culvert) than you are ok on this account.

The inlet losses (not to be confused with inlet control) for cross sections with lids is going to be computed based on the contraction loss, which has a default of 0.1. This should be higher. An additional confusion regarding the inlet loss is how it is computed. Unlike the exit loss, the “traditional” culvert equation bases the inlet loss on the actual velocity times the inlet loss coefficient. The standard cross section with lids/energy/backwater computation uses the difference in velocity between two cross sections. If the culvert causes a lot of constriction, this difference is minor. If there is not much constriction as the flow enters the culvert, the culvert equation will give a higher loss. This difference is usually pretty minor, but to get the exact same answer you could set a “k loss” value using the Set Internal Changes in WS and EG. This k loss will behave the same way as the inlet loss.

The culvert equations assume there will be a significant contraction as the flow enters the culvert. For a zero rise situation, this assumption is bad. Because of this, the inlet loss should probably be set to zero. However, you should check that the velocity in the culvert is no higher than the velocity immediatly upstream or else you may be zeroing out actual losses. A smooth transition (wingwalls/rounded edges etc) for the upstream end is also probably going to be important.

If you already know the maximum flow that the culvert is required to pass without exceeding some given headwater, then stick with steady flow. If the maximum flow is to be determined based on routing some given hydrograph, then you would want to use unsteady flow.

To get a zero rise will essentially mean that the inlet of the culvert will have open channel flow (as opposed to flowing full) and it will mean that the culvert is probably not under inliet control. If the culvert slope is steep enough, it would be possible for the downstream part of the culvert to be flowing full even though the entrance is not.

[Author]

Posts