Question on intent of software for Mr. Smart

[gadwall]

Mr. Smart,

Often in your posts you refer to keeping the model as simple as possible. Can you give a little more insight into "as simple as possible?"

My local municipality has adopted HydroCAD as its modeling software of choice, but the models that are created for say a residential subdivision quickly become complex. It is percieved by the Engineering community that the municpality is asking for the model to reference every drainage area, no matter how small, every catch basin and every pipe in the system. Is this a correct interpretation of what HydroCAD is intended to model?

 

[psmart]

There are no fixed rules, except to say that including elements in the HydroCAD model that do not significantly effect the results produces a model that is unnecessarily complex. This may make it harder to work with, and sometimes less stable.

Very complex models are certainly possible, and can operate properly. But the user needs to be clear on their modeling objectives before they start throwing every detail into the program.

Remember that at it's core, HydroCAD provides a set of hydrograph generation and routing operations. These operations (nodes) must be properly applied to the site in question. Each node may - or may not - correspond to a specific physical structure on the site. This translation requires:

1) A clear understanding of the modeling objectives
2) A good knowledge of the techniques being employed.
3) Sound judgment by a qualified designer

Without all of these elements you get the old garbage-in garbage-out scenario.

For general modeling guidelines please see

http://www.hydrocad.net/howto.htm

For more specific information on storm sewers (which I suspect may be at the heart of your question) please see

http://www.hydrocad.net/sewers.htm

There is lots more info on the general support page at

http://www.hydrocad.net/support.htm

 

[ctbailey]

Hydrocad is based on the SCS drainage methods. Many of the SCS items (such as Curve Numbers) have remain unchanged since their original inception, back in the 50's.

The SCS was charged with devising a method to estimate runoff, to prevent massive erosion like that which occured 20 years prior (the 1930's dustbowl.)

The numerical methods scientists and engineers had at their disposal were limited to tabular information, slide rule computations, and other "analog" items. (Of course, a skilled operator could perform long divsion carried out to the fourth decimal on a slide rule, but that had to be laborious.)

These SCS methods, the rational method, and then the modified rational method were originally designed in terms of square miles.

not hundreds of square feet!

It is my opinion that municipalities and other regulatory agencies that require designers to model subcatchments down to the nearest 100 square feet are abusing the software.

Can you trust an emperically derived curve number, one that was tested on ground cover in test plots of square miles, and linearly scale that down to the contributing area of a sinle catch basin?

Just because a PC can divide a number by 43,560 does not mean the answer is correct.

I suggest you keep your subcatchments as big as possible, because the runoff CNs are not appropriate for tiny subcatchments.

Other mis-use of the software:

I have recently had a reviewer ask me to determine the potentional pressure head of a 6" perforated underdrain that was designed to be draining a rock filled trench, because if the drain were to operate under a pressure head, the capacity of the underdrain would be limited, and then the rock filled trench would drain slower than the model indicates.

It would drain slower than the model indicates because HydroCAD can not handle a pipe flowing full, and is limited to open channel flow in pipes.

The manual is direct and upfront about this limitation, and suggests users apply the dynamic storage indication routing method, to include the tailwater effects on pipe systems when trying to model catch basins.

So now the model needs to model the tiny subcathments that contribute to individual catchbasins, the culverts connecting them with a tailwater analysis, and also a pond with insignificant storage at the daylights of underdrains.

How's that for keep the model simple?













___
Craig T. Bailey, PE

http://www.bailey-associates.com

 

[psmart]

ctbailey - These are legitimate issues that must be considered by any designer using any piece of software. HydroCAD is a collection of tools, and it is up the user to apply them as they see fit.

It's worth noting that there is a margin of error with all H&H techniques that cannot be eliminated. However, a model can do a much better job of projecting changes (from exiting to proposed conditions), than predicting exact values.

I must correct you on one point: HydroCAD CAN handle pipe/culvert flow with headwater and tailwater conditions, including full-flow conditions. What you are referring to is that the fact that a pipe REACH is based on open-channel flow and is not appropriate for pipe-full conditions. If the pipe flows full, the software issues a warning that advises the use of a pond/culvert instead. This WILL handle headwater, tailwater, and pipe-full conditions, and is clearly explained in the manual and help system. Once again, the user needs to understand each of the tools in the box and use it appropriately. For details on pipe modeling options please see

http://www.hydrocad.net/pipes.htm

Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[gadwall]

Craig,
I agree 100% with what you are saying. The problem is that the reviewing agencies are hiring younger engineers who lack experience. These reviewers, most of them not licensed, are trying to dictate how a PE should design his site. I think this is abominable, at best. When asked why they think something should be designed a certain way the answer is always "That's what the manual says." There is little or no use of common sense in the local municpalities that I work in anymore.