Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Roadside Ditch

Status
Not open for further replies.

peter246

Civil/Environmental
Mar 9, 2006
45
Hello, i'm doing a review of runoff calculations & noticed that the applicant hasn't set up the model as they described the actual conditions, and i've been asked how the shoulda/coulda done it. there are a series of subcats that flow to a continuous roadside ditch that has culverts and they've delineated the subcats from the culverts, so that each subcat flows to a culvert.

as the ditch is (presumably) continuous, and they have not described different, whatever can't flow through each culvert should cascade downstream through the ditch - this is how their Drainage Diagram is set up for routing, as dashed lines connect the nodes, but their output shows zero flow from the Secondary Outflow for all storms and each culvert, and impossibly high flows through the roadside culverts (54 cfs for an 18-inch CMP?! c'mon...)

please note that they did not set up the culverts as catch basins, and their output reports don't include details on how they set up any sort of storage, which they must have, otherwise HCAD won't let you save a pond node, right?

Anyway, at each design point (culvert), I think they should:
1) define the culvert as the primary outlet;
2) define a trapezoidal weir as secondary outlet at the same or slightly
higher invert as the culvert (the weir having the same geometry as the ditch); and
3) provide for overflow along the road through a broad-crested weir as tertiary outlet routed to same place as the culvert (it's a small ditch).

they also should use reaches in-between the culverts to represent the roadside ditch.

sound reasonable? not having done this particular physical setup before, i'd appreciate input. thanks in advance!
 
Replies continue below

Recommended for you

The high flow sounds like a high-head situation on the culvert, which can occur if the appropriate overflow device is not defined. Since their secondary outflow is zero, I would verify that the overflow device is setup properly. (Perhaps they set the discharge multiplier to zero?) I would have to see the actual HydroCAD file to comment further.



Peter Smart
HydroCAD Software
 
i agree, but i don't think we can get the model.

I’ve never seen a model that shows a secondary outlet connected to another node if no secondary outlet is defined… the option to connect the secondary outlet to another node doesn’t become available until a secondary outlet is defined. I was able to duplicate their setup in HydroCAD by defining a secondary outlet, routing it to another node, and then deleting the secondary outlet definition – the secondary outlet remained routed to the node that I chose despite having been un-defined in the program.

So i think they deleted the secondary outlet definition after routing it, or set the multiplier to zero like you said.

regardless, did i describe modeling the ditch/culverts correctly? i'd use CB's instead of ponds...
 
A CB is simply a pond with no storage, which tends to produce a more stable routing with very small "ponds" such as catch basins. If there's enough storage to cause noticeable peak attenuation, you should enter the storage, otherwise the storage has no effect and can be safely neglected.

As for the ditches and culverts, I'm not really clear if the culverts are in the main ditch line or on the contributing branches.

Too much speculation here - I really need to see the HydroCAD project file in order to have an informed comment.


Peter Smart
HydroCAD Software
 
i'd like to but can't discuss specifics publicly... i could PM you a link to the design docs & point out the specific sections to save you time searching through the 700+ pg SWPPP

my basic question isn't specific to their project - i've often thought about how to model the continuous ditch/intermittent culvert situation in HCAD, and know how i'd approach it, just haven't had a project to do so yet.
 
My .02, based on your description:

If they have that much flow going through an 18 inch culvert, they've probably got the culverts set at wrong (very low) elevations in their model, compared to the elevation of the ditch. And if that's the case, the water's all falling down an imaginary hole and going through the culverts, instead of some of it going further down the ditch.

You should be able to tell this from the output file, I would think, if they provided a complete printout of the routing calcs.

Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -
 
they didn't provide complete printouts, but enough to reconstruct the model with a little digging, which is the direction we're headed i think.

Beej, the problem is that, even though they appear to recognize that water would continue down the ditch, they don't define ANY outlets at the ditch culverts. as a result, they have, for example, 88 CFS being somehow taken care of through an 18" culvert, with no overtopping of the road and no flow continuing down the ditch to the next node THAT THE SECONDARY OUTFLOW IS GRAPHICALLY CONNECTED TO! (sorry for yelling ;)

i have to explain the situation to the atty's & want to be able to explain how it arguably should be done otherwise - i'm just trying to troubleshoot my technique. we may even get the chance (i.e. be paid to :) to reconstruct the model & play around with it to see what effect it has.

 
What do you mean by "continuous ditch/intermittent culvert"? Do you mean culverts set in the ditch at intervals, such as driveway culverts in a roadside ditch? If so, this is commonly modeled as a series of ponds with culvert outlets, and perhaps an overflow weir to handle any water that overtops each driveway. The pond storage would be the volume that is inundated as the WSE rises to the top of the driveway fill above the culvert. Both outlet devices (culvert and weir) would be routed to the next "pond" in the chain. Unless the ditch sections are very long, a interconnecting reach is not recommended since this would "break" the automatic pond-to-pond tailwater capability. (and the reach routing effects are minimal unless the reach is fairly long e.g. 1000 feet)

You can always contact HydroCAD support with any questions.

Peter Smart
HydroCAD Software
 
i wish they would have just used links instead of Pond:None nodes, and not "sneaked" in those secondary routing indicators* - made it look like they were modeling it different than they did...

* i still don't see how to do that without defining one, routing it, then removing it when editing the node...
 
Do you have the pond summary report? This will show the complete outlet setup for each pond. Also check the maximum WSE. If this is significantly above the overflow elevation for any pond it's likely they've omitted the overflow device.

Peter Smart
HydroCAD Software
 
no, not driveway culverts - ditch on uphill side of road, with culverts to drain the ditch to the downhill side.
road & ditch are 5 to 10% grade. distances between culverts range from ~300 to 700 ft. i don't believe there are check dams or submerged inlets to encourage water to the culverts.

i think i understand why they did it so simply - their design points end there (along the road), and because they've shown that peak flows will be attenuated to each point, they (obviously) don't feel the need to address the predicted flooding

i'm wondering how the hydrographs add up when properly routed, though... it's my contention that they can't do a valid pre- / post- comparison until the routing is set up correctly to model the time-varying response of the downstream nodes that should be receiving flows that the upstream culverts can't get rid of.
 
OK, now I understand the design. Steps 1,2,3 in your original post sound like a reasonable modeling approach, probably using a zero-storage pond. You could connect one pond directly to the next, or use an intervening reach if you feel the travel time will have a significant effect on the model. Just remember the a reach isn't tailwater-aware, so it will break the pond-to-pond tailwater capability that would otherwise be present.

Peter Smart
HydroCAD Software
 
good deal - thanks for the input! the ditch is so steep & it's so long in-between that i'm not worried about tailwater connections between the culverts inlets in the ditch. i think that the travel time could affect the calcs, but that's easy enough to investigate if we are asked.
 
Just tell them their design is obviously borked, and to either fix it or defend it by submitting a complete summary report so you can show them where it's borked.

Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -
 
the situation's a lil' more complicated, but i'm glad for the opportunity for discussion here - always learning... haha! 'borked' a new (to me) technical term!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor